New hardware & suPHP - Updates from the office

[admin]

Many of you have been asking about the upcoming rollout of suPHP. Well, I'm happy to announce we're getting ready to make final arrangements on the changes. What we did notice was that there was a performance issue with the current hardware and suPHP - that's why we haven't already gone ahead with the cut-over.

To that end, today we brought new hardware online with "uber" amounts of RAM, RAID5 storage and gigE ports. This new hardware was load tested and we're confident it's the right combination to handle suPHP, along with our continued growth.

We'll begin moving customers soon. The first batch of customers/servers will be notified in the coming days via the email address in your Directadmin control panel. (Be sure that's updated and current).

The good news - there's no IP changes. There may be a short period of downtime as we migrate your data to the new servers, but we're planning on a weekend/overnight change, so impact should be negligable. We'll post details as we near the first changes.

I'll be posting more on this in the coming days, along with a status report when the first servers go live.

Thanks everyone - we continue to grow with your help!

[admin]

After some major delays on my part due to health issues, I'm moving forward with this plan this weekend.

The new hardware is in place and I'm ready to start backing up the servers, one by one and moving them to the new hardware.

If you remember the last time we moved clients, it was a direct copy from one physical box to another. There was no data loss, but there was downtime associated with the transfer to insure we had all data, in tact - including databases and email, without a problem. I will employ the same strategy this time, with one notable exception. When the switch is flipped to the new server, it will be running suPHP. This MAY incapacitate some scripts, notably those that are currently chmod 777. suPHP will not allow chmod 777. You'll need to make these changes (probably to permission level 755) yourself, it's not something WE can do. If you're unsure how to do this, please check with the script author, or your FTP documentation.

I'm going to start with two servers this weekend. www520 and www534. These are currently our hardest hit servers and those most in need of a hardware upgrade. During the transfer, we will disable all services, mail, web and mysql - to make sure we're getting the latest data. Once everything is restored, services will be re-enabled.

If you have any questions, please ask them here for everyone's benefit.

I'll be emailing everyone on the affected servers as the move progresses. I'll be starting the backups on Sunday, approximately 2pm EST. I'm estimating 12 hours for all data to be moved and restored. There will be NO ip changes or DNS changes necessary.


Joe

[admin]

The backup process has begun on these two servers. Services have been temporarily disabled while the backups and transfers are happening.

I'll update more via this thread as we progress through the afternoon.

Thank you for your patience

Joe

[acm]

What happened to "I'll be emailing everyone on the affected servers as the move progresses."? I just discovered in a round about way that this was happening (all websites and email service down!).

[admin]

Emails were sent from the addresses in the Directadmin control panel prior to the transfer beginning.

[caddickj]

Emails were sent from the addresses in the Directadmin control panel prior to the transfer beginning.

Obviously there's been some oddity around here lately, which is concerning. As to this piece, though, I didn't receive an email (and I verified my DA email last night, received a backup automated notification early this morning, and I've checked my spam folder).

Has work been started on 534? I can't get to my blog, which has happened before when I've been moved from server to server.

Update: I realized I posted this at 2pm, when Joe had said the switch would start. Missed that earlier. Still... never got an email.

Also, came up with another question: is the switch to SuPHP likely to cause issues with WordPress? I know I've had to use 777 previously with that.

[acm]

Emails were sent from the addresses in the Directadmin control panel prior to the transfer beginning.

My email there is the one I've always used, it's still valid, and I opened a couple tickets this am re the down host and received emails on the open tickets to that address, so there is clearly something wrong with the notification process.

Did anybody receive a notification?

[admin]

Has work been started on 534? I can't get to my blog, which has happened before when I've been moved from server to server.


As I stated above, yes, the process has started:


I'm going to start with two servers this weekend. www520 and www534. These are currently our hardest hit servers and those most in need of a hardware upgrade. During the transfer, we will disable all services, mail, web and mysql - to make sure we're getting the latest data. Once everything is restored, services will be re-enabled.

[ozee]

Emails were sent from the addresses in the Directadmin control panel prior to the transfer beginning.

I got nothing.

Actually, I've received absolutely NOTHING during this entire process. Nothing that you were going to do anything, nothing that it had started...

When are you going to start telling us what's going on!!??

[admin]

My email there is the one I've always used, it's still valid, and I opened a couple tickets this am re the down host and received emails on the open tickets to that address, so there is clearly something wrong with the notification process.

Did anybody receive a notification?


The address in the directadmin control panel is set independently of the helpdesk system. They're two unique pieces of software. When the backup process is done, I'll verify this with the control panel mail logs.

All resources are focused on this backup this weekend, it's a major priority for us / me. It must happen, this is the only way to insure it happens in full, accurately and completely the first time, without screwups.

[ozee]

My email address is the same both in the helpdesk and in directadmin. I got nothing!

[admin]

Backups are still in progress, new server is ready to rock and roll once we get the backups transferred.

Going very well so far.

[Just J]

My email address is the same both in the helpdesk and in directadmin. I got nothing!

Same here. Something is not right.

[ssnapier]

I also got nothing in regards to this change. I used to subscribe to the hostpc.com twitter account until whoever was running it decided to post personal and political junk instead of useful, professional and timely information (which is what I was expecting).

I also have far too much going on to check in on this forum unless something is wrong (like now). There needs to be an effective solution in place because it is clearly not working right now. :mad:

[blob]

How is the move going? Any time frame on when you expect to complete it?
ps. I also have not received an email.

[admin]

Backups are still running, they're about 60% completed. This is one of the reasons for the move to the new hardware - disk access times are abysmal.

I originally scheduled 12 hours for the move - I can't provide any more of an ETA at this point.

I won't be able to check on the email logs until after this is complete, thanks for the info.

[esgdirect]

I received the following email at approx 12:30 Eastern Time today....albeit it was last minute.

We've begun the backup and transfer process of www520.hostpc.com (your web server) per this forum announcement:

http://hostpc.com/community/showthread.php?4299-New-hardware-amp-suPHP-Updates-from-the-office


Thank you for your patience.

Joe

[taylor]

I'm glad to see you back Joe. I hope your health continues to improve (or that you're already back to 100%). I have a couple questions:

1. Is the plan to move all servers to new hardware or there some servers that continue to handle their current load? If so, which ones won't be moved?

2. In the late 2009 server move thread (http://hostpc.com/community/showthread.php?4125-Server-migrations&p=20167#post20167) you wrote:


Thanks to new technology in this datacenter, we can do a complete recovery in MINUTES - from bare metal. It's really incredible - and another way we're doing our best to bring the best of the internet to you!
(end shameless sales pitch)

I would have expected it to be possible to use this technology to help out with these server moves. Is it no longer in place, or is there some other reason why it can't be used to perform a complete recovery on the improved hardware's bare metal?

TIA

[AuctionHugh]

All my domains are down. No notice, no clue, no email.

Happened up on this thread from a link from other threads.

I think this means that the reason all my domains are down is that some maintenance is being done on the servers - and they will be back up soon with no long term issues?

Can anyone confirm this?

[admin]

I'm glad to see you back Joe. I hope your health continues to improve (or that you're already back to 100%). I have a couple questions:

1. Is the plan to move all servers to new hardware or there some servers that continue to handle their current load? If so, which ones won't be moved?

2. In the late 2009 server move thread (http://hostpc.com/community/showthread.php?4125-Server-migrations&p=20167#post20167) you wrote:



I would have expected it to be possible to use this technology to help out with these server moves. Is it no longer in place, or is there some other reason why it can't be used to perform a complete recovery on the improved hardware's bare metal?

TIA


Yes - and no. MOST servers will move to the new hardware design, some are already there.

The bare-metal restores are exactly that, they copy bit by bit - and can only restore bit for bit. During this process however, it was decided, due to security issues, to make the servers utilize suPHP, now an industry standard. Converting from NON-suPHP to suPHP is a bastard (to put it bluntly) - this was the preferred method. Two birds, one stone.

We elected to do this on a Sunday, knowing it would take a considerable amount of time - to be the least interruptive to normal weekday traffic.

the backups continue - then the restore process. I have no further ETA - but we need to complete this at this time, one way or another.

Joe

[Howie999]

I also got the message, with only a few minutes to spare. This is quite unlike the legendary service I've gotten from HostPC in the years I've used them.

How about a status report? I can't afford this to be down any longer!

[edmicman]

Count me in as another one who didn't get the email. I've received automated emails from DA before, but nothing this time. Saw sites were down, came to the forum and determined this was the cause of it. I'm OK with it....upgrades are necessary and things like this are exactly why I moved my domain's email hosting to Google Apps a few years ago. All in all my stuff's mostly personal sites and doesn't really matter if it's down for awhile. But it's still kinda annoying to basically have to find out through the grapevine that there's something going on. I'll doublecheck my DA email settings once its all back up, but I'm pretty sure it was all working before.

Either way, good luck on the backups and restore and I'll be looking forward to better performance!

[admin]

I also got the message, with only a few minutes to spare. This is quite unlike the legendary service I've gotten from HostPC in the years I've used them.

How about a status report? I can't afford this to be down any longer!


Glad to see one person reporting getting the message.


the backups continue - then the restore process. I have no further ETA - but we need to complete this at this time, one way or another.


That's the best I can offer. I've been sitting here with the backups since about 12:30pm, tweaking new servers, etc. There's nothing we can do to make them go any faster - and I refuse to abandon the move at this point - it would make the last 12 hours of downtime pointless - only to have to do it again another day. It is what it is, we'll just need to let it complete as quickly as it can, that's the best we can do.

The extended backup times are a prime example of why we're getting off the older hardware - the drive array access times are horrible. The new servers are significantly better.

[ssnapier]

Mine is still down... what is the deal?

At about 8 pm you said it was 60% complete, we are now 9 hours past that and it is still down? This is frustrating to no end.... and now I have to go work (where hostpc is blocked) so not only am I going to be in the dark in terms of updates, if something goes wrong during this process I cannot access the control panel or my FTP to fix it. This is complete bull crap, and there better not be any problems when it comes back up or I will start looking for a new host.

[taylor]

Thanks for the quick and thorough reply Joe. Makes sense. I was just hoping that it would be possible to leverage the backup that must already be there. Obviously it's too late to do anything with these two servers now, but if you've got any more servers that are really struggling, would it be an option to do a bare metal restore to better hardware and then pull the backup off of the better hardware (presumably much quicker) and then use that backup to populate yet another server? ... just throwing out ideas here. I'm sure you've got a much better handle on what would or would not work.

[admin]

The backups are nearly completed - they're taking MUCH longer than we anticipated.

As soon as they're completed, we can move them to the new server, which should only take a few minutes thanks to the internal network, then we can begin the restores. Once the restores are started, we'll be able to move the IP's from the old servers to the new while it's restoring, and the accounts will begin to come back online in alphabetical order.

Please don't ask for an ETA, my original estimates have already been blown out of the water. I've been here since 12:30 yesterday babysitting these (with a 2hr catnap) - and I'll be here till it's complete. We're doing this the best and fastest way possible.

Thank you for your understanding.

Joe

[blob]

I am "t", I'll wager going live 1 pm :nod

[Howie999]

9 AM EDT Monday, 520 still down.

I've always found my site to have very fast response times on this "old" server; that's one of the main reasons I have NEVER considered another hosting company in the 7+ years I've been using HostPC.

I hope this outage doesn't scare away frequent visitors and/or new visitors. I'm getting pretty scared now.

Was there (or IS there) any way for your customers to upload OUR OWN BACKUPS of our sites to the new server? I would have gladly done that on my main sites.

Thanks, Joe! Hope this nightmare ends very soon.

[blob]

How about letting us know at what stage you are? Still on backup?

[AngryAnts]

chalk me up with the rest of the fellas. no warning email came to me either. after being down for 12 hours or so? came to the forums to see what was up.

how much data is there?

lol i forgot my username here.... nice.

[Carbo]

Like most everyone else here, I first found out about this when I realized my site was down and I had no access to my email. Never received an email or advanced warning of any kind. Bad form, guys.
I run a business website. If I'm down, I'm losing money. And with no advance notice being posted to my customers, I look bad. I can't tell you how this is hurting me and how disappointed I am with HostPC.
12 hours is now pushing a day and a half and counting. . .

[carbonspace]

I didn't get an email notification either. It would have been nice to accommodate for this instead of waking to a harsh reality this morning. :(:(:(

[badenovs]

Shoulda checked here first before creating ticket... and yes, after checking my Spam filter, I did get the message!

So, ETA on 520?


My email there is the one I've always used, it's still valid, and I opened a couple tickets this am re the down host and received emails on the open tickets to that address, so there is clearly something wrong with the notification process.

Did anybody receive a notification?

[atene]

I received an email from admin Sunday Morning just minutes (Ok - technically minutes could still equate to a year) before the server went off line. I have two sites on one of these servers and received a seperate email on both. Friday would have been nicer - but not much help to losing 24+ hours of emails . Hopefully the email server doesn't burn out delivering all the stored emails! :) You ARE storing emails - right? I will say I welcome tthe change - I was noticing for the previous week that email was timing out on send/receive for long periods of time, especially early morning (ie 5am CDT) And of course it couldn't happen at a worse time as a web designer was trying to test a 3rd site for client - not sure if they got an email or not.

[Just J]

Hopefully the email server doesn't burn out delivering all the stored emails! :) You ARE storing emails - right?

Things are not looking good. I just got a GMail from a friend with the rejection notice he received when attempting to send to my HostPC domain. It read in part:


Your message could not be delivered for more than 6 hour(s).
It will be retried until it is 1 day(s) old.


So:

HostPC is not capturing the emails but is rejecting them.
With a 24+ hour outage, it is possible many providers will abandon their attempts to deliver an email and you will never get it.

[Howie999]

Joe, I'm begging you. I think we all have been patient. But as has been pointed out, WE'RE LOSING MONEY.

1.) Can I upload my backup to the new server?
2.) WHEN can we expect our sites to be back up? I know your previous ETAs have been underestimating the problem, but having your customers in limbo for a day is frankly, just not acceptable.
3.) Is E-mail being cached, or being rejected - or is that a function of the sender?

[caddickj]

A friend of mine (another HostPC customer) has been unable to post here because he's waiting for admin approval on his new forum account. He asked me to post the following for him:


Can we at least get a status update? The last status update from Admin was at 6:29am EST stating that backups were nearly complete and it should be online soon, but the problem is... WE'RE NOT!

I'd appreciate another update myself. I'm hoping we're at least into the restore period at this point. Depending on what you're using for the alphabetical updates schedule, I'm either a 'c' or a 'j' so at best we're not even halfway through the restores yet.

Have there been unexpected problems, or is it just a matter of progress being slow?

Thanks.

[Sean]

Server moves are underway and are taking longer then anticipated. This is not like moving your desktop to a new desktop...there are a lot of "moving parts".

Please DO NOT open support tickets...they will simply be replied to with a link to this thread and closed. We do not have the staff or time to reply to dozens of tickets..thus the reason for the thread.

Also, this thread is for informational purposes, not for gripping...and as such will remain closed until further updates are available. Please understand that we are aware of everyone's concerns and Joe is working as fast as humanly possible. I'm not going to comment on Joe's health issue other then that we are glad to have in back and able to back. If Joe feels he wants to comment on it, that is his decision.

I know things have not been up to par (yes...that's probalby understating things), but hopefully we can start to get things under control again after these moves.

Sean

[admin]

The copies to the new server are approximately 50% completed, and we'll begin the restore process in chunks very shortly.

I'll post another update soon.

Joe

[atene]

Thanks Joe - I wouldn't want your headaches - even on a good day!

Take care and thanks for all you do.

Ray

[RocketRick]

Thanks Joe - I wouldn't want your headaches - even on a good day!

Take care and thanks for all you do.

Ray

Agreed 100%!

Despite the headaches that the users on those 2 servers are currently going through, I think that all of us are happy to see that you are on top of things, and still making the investment into server upgrades, etc.

Any plans for a similar hardware upgrade to 530? No pressure, but the performance lately has been somewhat sub-optimal, to put it lightly....

Thanks,

- Rick

[Howie999]

Is that a light at the end of the tunnel I see?

Or is that an oncoming train?

Joe, I truly appreciate your help and REALLY appreciate your updates. I was just about ready to jump ship when I saw Sean's comment (and the corresponding CLOSURE of this thread), because I thought that was the end. I think it's important for you guys to see how this affects your customers, so I'm glad the thread is back up again.

I've been with HostPC for 7+ years, and this has been by far the biggest hiccup in service, so in the grand scheme of things I can't complain too much. And I am looking forward to many more years of working together.

NOW can I have my site back up? ;)

[admin]

GOOD NEWS!

Restores have been started (PHEW!) - and are progressing without an issue so far. Shortly, the routes will beging re-converging to the new server so no IP changes will be necessary.

If you were affected by todays epic move (that I do NOT want to repeat) - you are on www520 OR www534.

Please - check your nameservers. They should be either:

ns520a.hostpc.com 67.231.253.4
ns520b.hostpc.com 67.231.253.5

OR

ns534a.hostpc.com 67.231.253.195
ns534b.hostpc.com 67.231.253.196

If you were affected today, and your DNS entries are something OTHER than these, they need to be updated to the correct server.

IP 67.231.253.3 is www520
IP 67.231.253.194 is www534

[admin]

GOOD NEWS!

Restores have been started (PHEW!) - and are progressing without an issue so far. Shortly, the routes will beging re-converging to the new server so no IP changes will be necessary.

If you were affected by todays epic move (that I do NOT want to repeat) - you are on www520 OR www534.

Please - check your nameservers. They should be either:

ns520a.hostpc.com 67.231.253.4
ns520b.hostpc.com 67.231.253.5

OR

ns534a.hostpc.com 67.231.253.195
ns534b.hostpc.com 67.231.253.196

If you were affected today, and your DNS entries are something OTHER than these, they need to be updated to the correct server.

IP 67.231.253.3 is www520
IP 67.231.253.194 is www534


Sorry, I hit submit too fast.

Restores are going in alphabetical order. Unfortunately, if you have a username starting in "z" - you'll be waiting a little bit longer, but it'll all happen before the end of the night.

Joe

[Sean]

For all you people who did not get the notification emails...our apologies, but think of it this way...

Your friend writes you a letter. They put a stamp on it. They go to the post office and mail it by handing it to the postal clerk who puts it in a bin for deliverly.

You never recieve the letter. Who is to blame? You? or the Post Office?

Maybe it was delivered but a gust of wind blew it away...
Maybe it was accidently delivered to someone else...
Maybe it got mixed up in the local catalogs and other junk mail...
Maybe your friend had the wrong mailing address and it was returned to them undelivered...

What you don't do is blame your friend, right?

Email works the same, people...it goes out in good faith that it will be delivered....unfortunately, thanks to the spammers out there, there are so many filters and other problems with the email system as it exists today that you are more likely to get that letter your friend mailed you via USPS then your email.

Sean

[badenovs]

Thanks for the updates... I will monitor this thread and stay off the ticket system (promise)!

And thanks for paying attention to us little guys!

[AngryAnts]

is 520 and 534 rebuilding concurrently?

I'm an "a" w00t! So i will post to this thread when I'm back online.

Thanks joe and team!

[acm]

GOOD NEWS!

Restores have been started (PHEW!) - and are progressing without an issue so far. Shortly, the routes will beging re-converging to the new server so no IP changes will be necessary.

If you were affected by todays epic move (that I do NOT want to repeat) - you are on www520 OR www534.

Please - check your nameservers. They should be either:

ns520a.hostpc.com 67.231.253.4
ns520b.hostpc.com 67.231.253.5

OR

ns534a.hostpc.com 67.231.253.195
ns534b.hostpc.com 67.231.253.196

If you were affected today, and your DNS entries are something OTHER than these, they need to be updated to the correct server.

IP 67.231.253.3 is www520
IP 67.231.253.194 is www534

All my nameservers are setup to us just ns534a.hostpc.com/ns534b.hostpc.com (no ip addresses) and they've been that way since before the last move or two ago, so I shouldn't have to change anything, right?

Now, when I ping my domains it says I on www534d.hostpc.com (67.231.253.197) and not www534.hostpc.com (67.231.253.194) and from what I remember It's been that way for a while. Is that correct?

[admin]

Restores are still progressing nicely - routes will begin to reconverge within the hour hopefully and we can get back to total stability.

Joe

[godde]

How do I know which one to use? :)

[badenovs]

Anybody's site up yet?

[AngryAnts]

mine is still down...

[MrK]

Your friend writes you a letter. They put a stamp on it. They go to the post office and mail it by handing it to the postal clerk who puts it in a bin for deliverly...

If I'm renting a condo from my friend and he writes me a letter informing me that he needs me to move out, goes to the post office and mails it by handing it to a postal clerk -- then drives directly to my place while I am away and changes the locks, I blame my friend.

[AuctionHugh]

I have one HostPC account with several domains hosted on it.

Some of the domains have the nameserver set to:
NS534A.HOSTPC.COM
NS534B.HOSTPC.COM

Others have it set to:
NS508A.HOSTPC.COM
NS508B.HOSTPC.COM

I only have one account, and all my sites are down (still down), so I am guessing I should change the 508 to 534 on them all. The thing is they all are working fine (up until yesterday) and I never received anything indicating the nameservers had changed or needed to change.

Anyway do I need to change all my 508 nameservers to 534 asap?


GOOD NEWS!

If you were affected by todays epic move (that I do NOT want to repeat) - you are on www520 OR www534.

Please - check your nameservers. They should be either:

ns520a.hostpc.com 67.231.253.4
ns520b.hostpc.com 67.231.253.5

OR

ns534a.hostpc.com 67.231.253.195
ns534b.hostpc.com 67.231.253.196

If you were affected today, and your DNS entries are something OTHER than these, they need to be updated to the correct server.....

[admin]

Yes AuctionHugh - all yours should be 534

[admin]

I've just been informed by the datacenter that in order to port the IP's over to the new hardware (which prevents any IP changes and DNS changes) the server needs to be rebooted with the correct IP's attached.

That, unfortunately, adds a couple hours to our transition - as all sites need to be fully restored (they're all in the que now) before the reboot can happen. Once it does, they'll all be online immediately.

We're currently 55% through the restore process - I'm honestly envisioning a 2am estimated completion time, but it could be sooner, this hardware is SCREAMING fast on the restores.

Anyway, they're pushing as fast as possible - I'll be here till they're completed. Tomorrow I'll be looking at mail logs to find out what happened to the notices I sent out concerning this - until then, I can't even begin to address it, but you have my word that I'll look into it and address any issue before another move is started for the remaining servers.

I'm going to grab a couple hour nap - a datacenter tech is monitoring the restores currently and will alert me if there's an issue. I'll be back to finish it up personally and insure everything went smoothly.

Thank you again for your patience. I'm sure we'll be able to work out a credit for the downtime you've experienced when this is completed.

Joe

[badenovs]

I"ll get some rest as well! Fwd'd your last post to my clients. I'll deal with them later!:nod

[Just J]

Tomorrow I'll be looking at mail logs to find out what happened to the notices I sent out concerning this - until then, I can't even begin to address it, but you have my word that I'll look into it and address any issue before another move is started for the remaining servers.

Please also look into the updating of the forums on this site, starting with the Server Status Reports forum, which used to have timely information about what's going on, and what's coming up. Instead, it features a sticky with instructions on how to sign up for EAS. I don't know whether to laugh or cry.

[AuctionHugh]

I changed the dns. Hopefully it will be good.

I am subscribed to this thread with instant email notification to my main yahoo email account. I never get those email notifications. This feature works fine with other forums. FYI.

[admin]

I changed the dns. Hopefully it will be good.

I am subscribed to this thread with instant email notification to my main yahoo email account. I never get those email notifications. This feature works fine with other forums. FYI.

Check your spam settings at Yahoo - that's the best I can offer right now.

[admin]

Ok folks - last update for tonight, I think.

The restores are currently in the p's - meaning everything before that has restored, and it's chugging out the last users as we speak. I'm ESTIMATING it'll be around 2am before the restores are done.

Now, the good news (knock wood) - I was able to wrangle some of the IP's on the server to start answering immediately. SOME sites are starting to resolve, I checked a small handful, they SEEM ok. I'm NOT saying it's done or perfect yet - but we are definitely making progress and there IS a light at the end of the tunnel.

I'll continue to do what I can while the CPU and RAM do the main part of the work. The drives (RAID 10) still need to completely sync up (essentially, copy all data to all drives) - but that'll happen transparently in the background.

Overnight, mail will start to flow back in as routes reconverge and DNS resolves. By morning, 99% of the data should be back online - we'll continue to work on that remaining 1% until it's completed 100%.

Right now, to save server loads, SPAM ASSASSIN is OFF - don't yell. Once the data is done restoring and things are stabilized, we will be turning that back on, but right now it's imperative to give 100% of the CPU to restoring and stabilizing.

Thank you again

Joe

[esgdirect]

I am back up - mail was first then sites about 15 minutes later.

[admin]

great news!

[Chuck Jamison]

I just received the notification now that the servers and sites are coming on-line. I post this, not as a complaint but rather to let you know that in your investigation of that notification process, it appears that the email was stranded somewhere.

I will change my notification email to my gmail instead of the one on this server.

Cheers

[AngryAnts]

My sites are slowly coming back online it seems. Oddly some of my subdomains on my domain work.. others not so much....

Will check in the AM. Thanks Joe

[ssnapier]

You apparently broke my forum with this "upgrade" so now not only did I wait the better part of 2 days for this crap it is STILL not working, this is seriously pissing me off now.... ticket inbound.

[blob]

looks like I am up

[Just J]

I'm up for web, but email is still down. Webmail throws an error about a missing preferences file. Guess I'll open a ticket.

[admin]

From the very first page in this thread (second post)


When the switch is flipped to the new server, it will be running suPHP. This MAY incapacitate some scripts, notably those that are currently chmod 777. suPHP will not allow chmod 777. You'll need to make these changes (probably to permission level 755) yourself, it's not something WE can do. If you're unsure how to do this, please check with the script author, or your FTP documentation.

We'll be dealing with individual issues one at a time today, until they're cleared.

[AuctionHugh]

Both my sites with forums, accessing the forums I get:

Internal Server Error
The server encountered an internal error or misconfiguration and was unable to complete your request

I believe I am running Simple Machines Forum, latest version.

I have no idea if that might be the chmod issue you mentioned.

Should I open a ticket on this with the specifics of the two domains?


Shortly, the routes will beging re-converging to the new server so no IP changes will be necessary.

If you were affected by todays epic move (that I do NOT want to repeat) - you are on www520 OR www534.

IP 67.231.253.3 is www520
IP 67.231.253.194 is www534

I have two sites where I do not control the nameservers and the owner is using an IP address to point to the sites instead. These were working fine before the moves by the way. The IP for these sites is 67.231.253.194.

These sites both return:
Apache is functioning normally
when I access them.

[taylor]

Hugh,

I'm guessing that this is a result of the chmod issue. You'll need to adjust the permissions so that they are not set to 777. If you have SSH access, you can do it from the command line, otherwise I think it has to be done either via the file browser in DirectAdmin or FTP (but I think those methods would be very tedious because AFAIK you have to do it for each file/subdirectory individually).

[admin]

Both my sites with forums, accessing the forums I get:

Internal Server Error
The server encountered an internal error or misconfiguration and was unable to complete your request

I believe I am running Simple Machines Forum, latest version.

I have no idea if that might be the chmod issue you mentioned.

Should I open a ticket on this with the specifics of the two domains?



I have two sites where I do not control the nameservers and the owner is using an IP address to point to the sites instead. These were working fine before the moves by the way. The IP for these sites is 67.231.253.194.

These sites both return:
Apache is functioning normally
when I access them.

Yes, my guess would be a chmod issue.

If they're using the IP's, it may need to be updated to the new master server IP - 67.231.248.250

[admin]

Hugh,

I'm guessing that this is a result of the chmod issue. You'll need to adjust the permissions so that they are not set to 777. If you have SSH access, you can do it from the command line, otherwise I think it has to be done either via the file browser in DirectAdmin or FTP (but I think those methods would be very tedious because AFAIK you have to do it for each file/subdirectory individually).


You can do this recursively through the file manager - or recursively via most standard FTP programs

[AuctionHugh]

You can do this recursively through the file manager - or recursively via most standard FTP programs

When I installed the forums for these sites I used the installatron script, likewise to keep it them updated. I'm not sure what files or folders would need to be chmoded to 755.

EDIT: add, I would appreciate help with this, and the below.


If they're using the IP's, it may need to be updated to the new master server IP - 67.231.248.250

I can have the person who manages the domain names change this if I know this is definitely true. Last time IPs changed we were given a couple weeks notice and very specific information so our sites would not be down. I don't mean to be pissy about this but if I tell this guy to change it and then have him change it back because that is not the issue I will look worse than I do now for having their site down for 2 days (plus propagation of new IP).

Do I definitely ask this guy to change the IP?

[AngryAnts]

You apparently broke my forum with this "upgrade" so now not only did I wait the better part of 2 days for this crap it is STILL not working, this is seriously pissing me off now.... ticket inbound.

I just logged in here to see if it happened to anyone else. I'm glad!

SMF right?

Check out this brilliant HPC response when I submited a ticket last night about my forums being down:


Hello Anthony,

I checked the subdomain http://xxxxx.antsxxx.com, and as you say it shows a 500 Internal Server Error. Please can you check the requirements on server for your forum, so that we can assist you with it.

Thank You,

Roy
HostPC Support

Really? Man... can someone who knows what theyre doing PLEASE look into this.

[admin]

For users affected ONLY: If anyone is comfortable updating your DNS - you can update them where you registered your domain

ns550a.hostpc.com
ns550b.hostpc.com

Thank you

[admin]

When I installed the forums for these sites I used the installatron script, likewise to keep it them updated. I'm not sure what files or folders would need to be chmoded to 755.

EDIT: add, I would appreciate help with this, and the below.



I can have the person who manages the domain names change this if I know this is definitely true. Last time IPs changed we were given a couple weeks notice and very specific information so our sites would not be down. I don't mean to be pissy about this but if I tell this guy to change it and then have him change it back because that is not the issue I will look worse than I do now for having their site down for 2 days (plus propagation of new IP).

Do I definitely ask this guy to change the IP?


You have private DNS setup at GoDaddy - yes, it'll need to be updated to ns550a.hostpc.com / ns550b.hostpc.com

[Chuck Jamison]

My site shows a message: Apache working (or something like that). Do I have to change pointers from 520 to 550? I'm www.2jamisons.com

Dreamweaver was able to update files but I can't get to my DA - my user name/password combination no longer work and when I used lost password, I get nothing by return email. (BTW email seems to be working but lots of spam and viruses).

[badenovs]

Your site seems to be up where I'm browsing from (Roadrunner, Upstate NY). Perhaps you need to refresh your DNS on your computer?

[acm]

A few observations:

- All my websites seem to be working now. One script was broken because a tmp directory for log files was missing, but I manually recreated it and now that is working. Why was it missing? Was that done intentionally by the backup/restore process or is it a sign of bigger problems?

- ssh works (ie that's how I fixed the problem mentioned above)

- Can't get to the DirectAdmin control panel, browser says "The connection has timed out", "The server ... is taking too long to respond."

- Can't access email (normally use POP on port 995 for SSL/TLS), email client says "Connection to server timed out."

- Using ssh I was able to look into the imap folders and see that email is being received without any delay. And, yes SPAMASSASSIN is not running

- The most import thing is that it now appears I am on www550 (67.231.248.250) and www534. Didn't Joe say they wouldn't be any IP address changes? Maybe he was just referring to the DNS servers?

[admin]

As I've mentioned in previous replies, we are NOT completed with this issue yet - there are multiple small issues and cleanup work to be done.

Our staff is working on tickets as quickly as possible. Please, do not feel like this is an "abuse the helpdesk" day - we understand the issues, we're working to clear them up as quickly as possible, and we realize the inconvenience that this move caused. We can't allow our staff to be abused verbally, threatened, harassed or spoken to with obscenities. It will not be tolerated.

We're clearing issues as quickly as possible. If you have an issue with this move, please feel free to open a polite, calm ticket, with as many details as you can provide, we'll get to it as quickly as possible.

Thank you for your patience and cooperation while we finish this up.

Joe

[AuctionHugh]

I'm sorry but I'm not clear on this. Where does the ns550a/b come into play?



If you were affected by todays epic move (that I do NOT want to repeat) - you are on www520 OR www534.

Please - check your nameservers. They should be either:

ns520a.hostpc.com 67.231.253.4
ns520b.hostpc.com 67.231.253.5

OR

ns534a.hostpc.com 67.231.253.195
ns534b.hostpc.com 67.231.253.196

If you were affected today, and your DNS entries are something OTHER than these, they need to be updated to the correct server.

IP 67.231.253.3 is www520
IP 67.231.253.194 is www534

[admin]

A few observations:

- All my websites seem to be working now. One script was broken because a tmp directory for log files was missing, but I manually recreated it and now that is working. Why was it missing? Was that done intentionally by the backup/restore process or is it a sign of bigger problems?

- ssh works (ie that's how I fixed the problem mentioned above)

- Can't get to the DirectAdmin control panel, browser says "The connection has timed out", "The server ... is taking too long to respond."

- Can't access email (normally use POP on port 995 for SSL/TLS), email client says "Connection to server timed out."

- Using ssh I was able to look into the imap folders and see that email is being received without any delay. And, yes SPAMASSASSIN is not running

- The most import thing is that it now appears I am on www550 (67.231.248.250) and www534. Didn't Joe say they wouldn't be any IP address changes? Maybe he was just referring to the DNS servers?


The firewall was preventing port 587 access, that should be resolved.
Spamassassin will be the last thing enabled.

How are you accessing Directadmin - I'm able to get in without any issue, quicker than ever.

[acm]

The firewall was preventing port 587 access, that should be resolved.
Spamassassin will be the last thing enabled.

How are you accessing Directadmin - I'm able to get in without any issue, quicker than ever.

Same as always: https://www.willowbunchtelegraphoffice.org:2223/

[acm]

The firewall was preventing port 587 access, that should be resolved.

I can send mail now, but still can't receive. As previously mentioned I use POP on port 995 for SSL/TLS. Might the firewall be blocking access to port 995?

[admin]

I can send mail now, but still can't receive. As previously mentioned I use POP on port 995 for SSL/TLS. Might the firewall be blocking access to port 995?

port 995 is not active - there's no SSL on your account at this time.

[admin]

Same as always: https://www.willowbunchtelegraphoffice.org:2223/


use http and port :2222

[Carbo]

My site is up and running quick and fine. Thank you.
I did have a glitch with my domain based emails, however. It's resolved now so maybe by posting here it can help someone else:
For some reason I was unable to send emails. I could, however, receive them. So I switched the outgoing mail port from 587 to the default port, 25. That seems to have solved the sending issue, as mail is now flowing in both directions.
Being a technology dummy, I don't know if I did the right thing or not, but as I said, it's working at this time.

[acm]

port 995 is not active - there's no SSL on your account at this time.

I expected that as it seems to happen every time we move. Will SSL be re-enabled soon or do I have to open a ticket for that?

[admin]

It's on the list

[admin]

SpamAssassin is being enabled at this time

[Chuck Jamison]

Your site seems to be up where I'm browsing from (Roadrunner, Upstate NY). Perhaps you need to refresh your DNS on your computer?

Thank you, thank you. I reloaded and I'm good now.

Looks like we have the AOL and Earthlink black list problems again. (Friends don't let friends have AOL accounts)

[admin]

Thank you, thank you. I reloaded and I'm good now.

Looks like we have the AOL and Earthlink black list problems again. (Friends don't let friends have AOL accounts)


Another staff member is dealing with AOL and Earthlink (same company actually) - they're getting the IP's whitelisted as we speak - then it's just up to AOL to propogate the changes

[admin]

Spamassassin is on hold till later today.



Directadmin via https has been updated to port 2223 (as before)

[caddickj]

well, my site's back, but I'm seeing performance instability (sometimes responsive, sometimes reeeeaaally slow, sometimes unreachable). For now I'm gonna chalk that up to you guys working on the server and hope it smooths out later.

[admin]

well, my site's back, but I'm seeing performance instability (sometimes responsive, sometimes reeeeaaally slow, sometimes unreachable). For now I'm gonna chalk that up to you guys working on the server and hope it smooths out later.


It will get significantly better - we're up to about 99.8% complete with everything - once that's done, everything will get back to "better than" normal.

Joe

[tonydi]

Are these two new servers virtual machines or real hardware? I ask because the specs reported by DA (dual Xeon E5504's, 8GB memory) are the same as the other existing servers.

[AngryAnts]

Did we just go down again?

[AuctionHugh]

Has anyone figured out how to get their SMF forum back up and running again?

My client's forum is dead in the water as of now.

Thanks.

[edmicman]

I'm sorry but I'm not clear on this. Where does the ns550a/b come into play?

This, too. My stuff seemed to be back up this morning, had a few Wordpress issues I noticed that I figured was the chmod issue that I'd fix when I got home. I didn't think I needed to adjust the DNS settings? Do ?

I'm on 534.

[gscott]

This, too. My stuff seemed to be back up this morning, had a few Wordpress issues I noticed that I figured was the chmod issue that I'd fix when I got home. I didn't think I needed to adjust the DNS settings? Do ?

I'm on 534.


No you shouldn't have to as long as the hot/swap is working
ns550a and ns550b are the nameservers (DNS) the complete address is ns550a.hostpc.com and ns550b.hostpc.com

If you want to move to the new Nameserver and that is actually a good idea you will need to login to your Domain Registrar account with GoDaddy or whomever your domains were registered to and change the Nameserver lines to the 2 new ones as long as that is the correct one for your domain.

C

[gscott]

Has anyone figured out how to get their SMF forum back up and running again?

My client's forum is dead in the water as of now.

Thanks.


Haven't trouble shot this myself but if 609 moves then I'll have to decide what to do with my SMF forum. . . This forum post at the SMF forums might help if you follow the directions:
http://www.simplemachines.org/community/index.php?topic=400017.0;prev_next=prev#new
(http://www.simplemachines.org/community/index.php?topic=400017.0;prev_next=prev#new)

[atene]

Does this need a ticket at this point? ATT blocking for abuse

host aln-mailrelay.att.net [12.102.252.75]: 521-67.231.248.250 blocked by sbc:blacklist.mailrelay.att.net.
521 DNSRBL: Blocked for abuse. See http://att.net/blocks

[AuctionHugh]

Has anyone figured out how to get their SMF forum back up and running again?

My client's forum is dead in the water as of now.

Thanks.

Changing my 777's to 755 in the directadmin file manager worked to bring the forum back up. It was slow as a sick dog to do though because the refresh on the directadmin file manager is incredibly slow. Might be because other maintenance issues are going on right now, but it was very unpleasant.

I did start a thread over at SMF where I am receiving some assistance and you may want to refer to it.

http://www.simplemachines.org/community/index.php?topic=401715.msg2792540#msg2792540

Note I did not mention that it was HostPC that is giving me unbillable headaches :nod

[Just J]

Anyone having security/certificate issues?

I went to check my IMAP email from an iPhone 4 - no response.

When I do if from my iPad (which is iOS V3.2 instead of iOS4.2), I got the warning that the certificate might be invalid. When I said to continue and forced the refresh, I could see all my mail.

[cutter44]

...I figured I might find a simple answer here.

After having email and my sites down since Sunday, I've been checking back here for answers (I didn't receive a warning email, either). I have a simple site. One main domain and about five sub-domains, all of which are giving me the Apache is functioning normally message when the page loads, and nothing else. None of my email accounts are sending or receiving. I'd be happy to change my DNS if I was sure I am supposed to, but I can't seem to find a straight answer and I don't want to change them and wait for them to propagate before I know if I'm supposed to. Right now I am using ns534a.hostpc.com and ns534b.hostpc.com. If I'm supposed to change them, what should I change them to?

While I'm sure it's aggravating for Joe and the crew, it is incredible frustrating for me, as I know nothing about fixing these issues beyond simple things like changing the DNS.

After three days down, I would just like people to be able to get to my sites and for me to be able to send and receive email.

[taylor]

Joe can correct me if I'm wrong, but I'm pretty sure that it will be no worse to switch your DNS to the ns550a and ns550b servers instead. The idea behind the hot-wiring is that the old (ns534a and ns534b) now point to the new server, so, if the hot-wiring is working, there is no need to change the nameservers (but it won't hurt).

Although this hasn't been said explicitly, it looks like all of the accounts on both www520 and www534 have been moved to www550 (which presumably has hardware that more than makes up for the increased number of accounts). Therefore, ns520a, ns520b, ns534a, ns534b, ns550a, and ns550b should all be pointing to the new server.

[admin]

...I figured I might find a simple answer here.

After having email and my sites down since Sunday, I've been checking back here for answers (I didn't receive a warning email, either). I have a simple site. One main domain and about five sub-domains, all of which are giving me the Apache is functioning normally message when the page loads, and nothing else. None of my email accounts are sending or receiving. I'd be happy to change my DNS if I was sure I am supposed to, but I can't seem to find a straight answer and I don't want to change them and wait for them to propagate before I know if I'm supposed to. Right now I am using ns534a.hostpc.com and ns534b.hostpc.com. If I'm supposed to change them, what should I change them to?

While I'm sure it's aggravating for Joe and the crew, it is incredible frustrating for me, as I know nothing about fixing these issues beyond simple things like changing the DNS.

After three days down, I would just like people to be able to get to my sites and for me to be able to send and receive email.

By all means, open a ticket

[admin]

Joe can correct me if I'm wrong, but I'm pretty sure that it will be no worse to switch your DNS to the ns550a and ns550b servers instead. The idea behind the hot-wiring is that the old (ns534a and ns534b) now point to the new server, so, if the hot-wiring is working, there is no need to change the nameservers (but it won't hurt).

Although this hasn't been said explicitly, it looks like all of the accounts on both www520 and www534 have been moved to www550 (which presumably has hardware that more than makes up for the increased number of accounts). Therefore, ns520a, ns520b, ns534a, ns534b, ns550a, and ns550b should all be pointing to the new server.


You're essentially correct. The hardware behind the new server is more than enough to handle the loads (which are running VERY lean right now, less than either server alone previously).

Anyone getting "Apache is functioning normally" - likely has a DNS issue.

Anyone getting 550 errors, should remove any .htaccess file they have, be sure directories and files aren't chmod 777 and go from there. If you still can't resolve it, open a ticket and we'll look closer.

We currently have two sites that we're really stuck figuring out, other than that, most of the issues are cleared up. There's a few lingering issues and last minute cleanups (like the aol/att/earthlink mail block) and spamassassin - but they'll be cleared in the next 24 hours, hopefully

[mrspike55]

Joe,

Having been through the whole AOL, AT&T, etc RBL issue in the past, wouldn't it be a good idea to submit the new IP ranges you plan on using BEFORE doing these type of moves?

Not being sarcastic, etc, just thinking!

[admin]

Joe,

Having been through the whole AOL, AT&T, etc RBL issue in the past, wouldn't it be a good idea to submit the new IP ranges you plan on using BEFORE doing these type of moves?

Not being sarcastic, etc, just thinking!

I submitted them over a month ago. Don't ya just LOVE aol?

[admin]

Due to dozens of customers on these servers not understanding how to CHMOD a file to 755 - and the number that had vulnerable scripts, configurations, etc - we're running a server wide change to ALL files, ALL directories chmod 755.

This will take about an hour to complete.

[Carbo]

Just wanted to chime in and say that all is well on my end. Site is up and running fast and smooth, including my IPB board. Emails are fine, also. As I posted earlier in this thread, I switched the outgoing mail port from 587 to the default port, 25. Once I did that, I was then able to send email, which was the only issue I was having once I was back online.
The one thing I am dealing with, though, is an unbelievable amount of spam flooding my inbox. Anyone else? What's the fix?

[admin]

We'll have SpamAssassin on board later today

[Carbo]

We'll have SpamAssassin on board later today
Thank you!

[ozee]

We'll have SpamAssassin on board later today


Is it "later today" yet?? Not on my accounts...

[admin]

Hey, gimme a break - there's still 2 hrs left on the east coast :) I'm working on it :)

[admin]

I'm watching logs extensively tonight, in between tickets, and I'm seeing quite a few MySQL login errors.

One issue I've found is you may need to reset the password for scripts in your Directadmin control panel, under "Mysql management" to match the ones in your setup or config files. If you're having an issue connecting to the mysql server, try that first please

[ozee]

Is it "later today" yet? Still not on my accounts...

Also, I'm still seeing the "Server fault" error messages and timeouts that I was seeing last week -- not as often, but still too often. The system still isn't "right", IMO.

[Howie999]

All my sites (6?) seem to be up and flyin' 100%. Spam still coming through by the bushel, but hey, I have an itchy delete finger, so that's not a BIG problem.

I seem to notice a definite speed increase (although I always thought server response was already pretty quick). Thanks for that, Joe and crew.

Now all I have to do is get some visitors back (I seem to have lost 50% of traffic). Fickle users who can't get on one day and then forget?

[hostpcuser]

For some people getting the 500 Server Error then it may be because of your .htaccess file. If there are any php injections into it (such as register_globals) then suPHP does not recognize and will give a server error.

HostPC through a ticket said to produce a php.ini in the public_html folder with the appropriate changed settings, yet it doesnt work that way. I have an open ticket trying to get this resolved so that their advice of creating a php.ini to change php settings, instead of through .htaccess.

Please reply if someone here knows how to do this as it has been over 12 hours since submitting the "high priority" ticket through HostPC ticketing system to get this resolved.

[jdiff]

For some people getting the 500 Server Error then it may be because of your .htaccess file. If there are any php injections into it (such as register_globals) then suPHP does not recognize and will give a server error.

HostPC through a ticket said to produce a php.ini in the public_html folder with the appropriate changed settings, yet it doesnt work that way. I have an open ticket trying to get this resolved so that their advice of creating a php.ini to change php settings, instead of through .htaccess.

Please reply if someone here knows how to do this as it has been over 12 hours since submitting the "high priority" ticket through HostPC ticketing system to get this resolved.

I had 500 Server Errors until I removed all the PHP directives in .htaccess.

Don't worry about php.ini files just yet. First, try renaming your .htaccess to htaccess.txt to disable it completely. If you site works (or at least works somewhat) then you can be pretty sure something in there was a problem. In mine, I had to remove all the lines with php_value and php_flag. If that doesn't work, comment out all the lines, then uncomment them one by one until you find the culprit.

[MrK]

HostPC through a ticket said to produce a php.ini in the public_html folder with the appropriate changed settings, yet it doesnt work that way. I have an open ticket trying to get this resolved so that their advice of creating a php.ini to change php settings, instead of through .htaccess.


I had the experience of converting a site to suPHP at another host. In addition to the php.ini file in the public_html folder, I needed to add the following line to the .htaccess file:
suPHP_ConfigPath (full path to public_html folder)
Example:
suPHP_ConfigPath /domains/example.com/public_html

I had a couple of AddType statements in the .htaccess file for processing .css and .js files as php files. Those statements were left in the .htaccess file.

[ozee]

Is it "later today" yet, Joe???

SA is STILL not working on my domains! I've submitted a ticket, but was only sent a form-letter reply telling me how to configure it. The problem is - it is NOT responding to my attempts to configure it! I make the changes, press SAVE -- and nothing has changed. It's telling me that it is "enabled", but there are no SA lines added to the email headers and there's as much (or more) spam as ever.

[esgdirect]

Spam assassin is enabled, but I am still getting whacked with a ton of Spam. I don't know a lot, but a thread that could offer suggestions on Spam Assassin setup and help with any plug-ins we could use would be great.

I have my custom threshold set at 2.5 and have added a lot of domains to the blacklist. I am using Thunderbird and have all the settings for that on and I just don't know what to do anymore.

I hate to open a ticket, but just give us some direction here on how to deal with this Spam. Please.

[acm]

Spam assassin is enabled, ...

How have you determined that?

Like ozee I still don't SA lines (ie X-Spam-...) on incoming email.

[esgdirect]

How have you determined that?

Like ozee I still don't SA lines (ie X-Spam-...) on incoming email.

During the server move, it wasn't installed. Then as of 2 nights ago I noticed is was installed, enabled and with my original settings still there. I know what your saying about the SA Lines. I am not a pro here...maybe I'm wrong but....

[admin]

There have been changes to the Directadmin release of Spamassassin which have crippled the binaries we have from performing correctly.

We're working to re-establish this as quickly as possible.

[hostpcuser]

I had the experience of converting a site to suPHP at another host. In addition to the php.ini file in the public_html folder, I needed to add the following line to the .htaccess file:
suPHP_ConfigPath (full path to public_html folder)
Example:
suPHP_ConfigPath /domains/example.com/public_html

I had a couple of AddType statements in the .htaccess file for processing .css and .js files as php files. Those statements were left in the .htaccess file.

This doesnt seem to work at hostpc. I've created my own custom php.ini in the public_html with register_globals = On in it and it doesnt turn it on. I believe that the issue is that the suPHP_ConfigPath in the .htaccess doesnt work.
I've tried the following with no success:

suPHP_ConfigPath /home/<userremoved>/domains/<domainremoved>/public_html
suPHP_ConfigPath /home/<userremoved>/domains/<domainremoved>/public_html/php.ini
suPHP_ConfigPath /home/<userremoved>/domains/<domainremoved>/public_html/
suPHP_ConfigPath /domains/<domainremoved>/public_html
suPHP_ConfigPath /domains/<domainremoved>/public_html/php.ini
suPHP_ConfigPath /domains/<domainremoved>/public_html/

Still nothing. Even after explaining the situation the response from the HostPC ticket said to put that php_flag statement to enable register_globals in the .htaccess, which we all know with suphp (and was explained in the original ticket to them) will cause a 500 internal server error.

Can someone help?

[hockey-dad]

My websites haven't converted yet so everything is still good for me. However in planning ahead for the possible disaster to come I found a couple of pages that seem to be very informative about how suPHP changes things. Hopefully they can help somebody.

https://support.apthost.com/index.php?_m=knowledgebase&_a=viewarticle&kbarticleid=716

http://drupal.org/node/139447

These may or may not be helpful but worth a shot.

[hockey-dad]

This doesnt seem to work at hostpc. I've created my own custom php.ini in the public_html with register_globals = On in it and it doesnt turn it on. I believe that the issue is that the suPHP_ConfigPath in the .htaccess doesnt work.
I've tried the following with no success:

suPHP_ConfigPath /home/<userremoved>/domains/<domainremoved>/public_html
suPHP_ConfigPath /home/<userremoved>/domains/<domainremoved>/public_html/php.ini
suPHP_ConfigPath /home/<userremoved>/domains/<domainremoved>/public_html/
suPHP_ConfigPath /domains/<domainremoved>/public_html
suPHP_ConfigPath /domains/<domainremoved>/public_html/php.ini
suPHP_ConfigPath /domains/<domainremoved>/public_html/

Still nothing. Even after explaining the situation the response from the HostPC ticket said to put that php_flag statement to enable register_globals in the .htaccess, which we all know with suphp (and was explained in the original ticket to them) will cause a 500 internal server error.

Can someone help?

Using Google, I found an older HostPC thread about how to set up the config path. There are 2 posts in the thread. The config path in the first one appears to be incomplete. Scan down the thread to the second post and that would appear to be the correct path string.

http://clients.hostpc.com/community/showthread.php?3991-Introduction-to-suPHP&p=19673

Again, my site isn't broken yet so I can't test this info :)

[MrK]

This doesnt seem to work at hostpc. I've created my own custom php.ini in the public_html with register_globals = On in it and it doesnt turn it on. I believe that the issue is that the suPHP_ConfigPath in the .htaccess doesnt work.


I did some testing and could not get suPHP_ConfigPath to work either. Could an admin check in /etc/httpd/conf/extra/httpd-suphp.conf and make sure that the suPHP_ConfigPath line is commented out. According to this article (http://help.directadmin.com/item.php?id=183), it needs to be in order for custom php.ini files to work.

[ricej]

Mail still blocked to AOL and add sbcglobal.net to the list.

[admin]

Mail still blocked to AOL and add sbcglobal.net to the list.

Complaints about mail to aol / att /sbcglobal should be directed to them. We've requested whitelist status a month ago, and twice this week - still no reply.

[kkip]

I did some testing and could not get suPHP_ConfigPath to work either. Could an admin check in /etc/httpd/conf/extra/httpd-suphp.conf and make sure that the suPHP_ConfigPath line is commented out. According to this article (http://help.directadmin.com/item.php?id=183), it needs to be in order for custom php.ini files to work.

I don't know if this will work, my HostPC server hasn't been converted. But I am using the following .htaccess code on another host with suPHP. It allows a local php.ini file located in the public_html directory to be used by scripts in all sub directories.



"<"IfModule mod_suphp.c">"
suPHP_ConfigPath /home/USERNAME/public_html
"<"Files php.in"i>"
order allow,deny
deny from all
"<"/Files">"
"<"/IfModule">"

Remove all quotation marks (can't post without them).

This code goes in the .htaccess file located in the public_html directory. Be sure to change the USERNAME to the atcual user name for the account.

[admin]

On first glance, the path needs to be modified also


suPHP_ConfigPath /home/USERNAME/public_html


should be:

suPHP_ConfigPath /home/USERNAME/domains/DOMAINNAME/public_html
</pre>

[MrK]

Complaints about mail to aol / att /sbcglobal should be directed to them. We've requested whitelist status a month ago, and twice this week - still no reply.

Are you kidding??? You've really stopping working on this? If you've experienced difficulty in dealing with these ISPs, imagine what it would be like for a single individual that does not have the resources and experience that you do.

[MrK]

I don't know if this will work, my HostPC server hasn't been converted. But I am using the following .htaccess code on another host with suPHP. It allows a local php.ini file located in the public_html directory to be used by scripts in all sub directories.



"<"IfModule mod_suphp.c">"
suPHP_ConfigPath /home/USERNAME/public_html
"<"Files php.in"i>"
order allow,deny
deny from all
"<"/Files">"
"<"/IfModule">"

Remove all quotation marks (can't post without them).

This code goes in the .htaccess file located in the public_html directory. Be sure to change the USERNAME to the atcual user name for the account.

I tried this using Joe's suggested path and it did not work. phpinfo() still lists the Loaded Configuration Path as: /usr/local/etc/php5/cgi/php.ini

Joe, was someone able to check for a suPHP_ConfigPath item in /etc/httpd/conf/extra/httpd-suphp.conf?

[atene]

Are you kidding??? You've really stopping working on this? If you've experienced difficulty in dealing with these ISPs, imagine what it would be like for a single individual that does not have the resources and experience that you do.

I have to agree. Everything worked before you upgraded your server. This was an issue last time you changed servers for me also. We suffered thru the process starting miday on a sunday plan, in hind sight - major mistake. Not testing transfer rates? We don't know what was or wasn't tried as there was almost no communication - only replies to pleads for corrections. Spam due to a DA upgrade? If mail is blocked because a Hosting companies IP range is blacklisted, that's not an individual issue, it is a hosting company issue. As a consumer everything with this upgrade was botched. Down time, lack of good clean proper Communication pre during and post. If I have to use Gamil inorder to email AOL or ATT how is HostPC meeting my needs? They aren't. A hosting plan isn't a bargain if it doesn't work, and yours isn't working for a whole lot of people. Your website provides a link to the DA features page which cotains features not available. http://www.directadmin.com/features.html front page extensions. Catchall. 2 that jump out at me. Its very frustrating we were left in the dark, and now apparently abandonded to fix on our own what wasn't broken one week ago today without notice. The ONLY email I received about this change was after it was too late.
11:43 am CDT We've begun the backup and transfer process of www520.hostpc.com (your web server) per this forum announcement:

http://hostpc.com/community/showthread.php?4299-New-hardware-amp-suPHP-Updates-from-the-office
That is not communication. Please bring back the minimum level of support. Where are all the posts? Have they all jumped ship?

[MrK]

I have to agree. Everything worked before you upgraded your server. This was an issue last time you changed servers for me also...

That's not quite true. Everything worked before EAS was dropped. After that, the additional load on the server started causing email to be rejected. I started experiencing periods of 2-3 hours a day in which email could not be received. It would be rejected on the sending end with the following error:

An error was detected while processing a file of BSMTP input.
The error message was:

421 Lost incoming connection

The SMTP transaction started in line 0.
The error was detected in line 3.
0 previous messages were successfully processed.
The rest of the batch was abandoned.
421 Lost incoming connection
Transaction started in line 0
Error detected in line 3

This whole server move has been a giant CF, but it was the sudden, unexplained drop of EAS that started it all off.

[acm]

Looks like SpamAssassin came back on-line in the middle of the night, here's what I see in my email headers this morning:


X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on www550.hostpc.com

This of course is an upgrade to what we had before the server move:


X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on www534.hostpc.com

Thanks guys!

[admin]

Looks like SpamAssassin came back on-line in the middle of the night, here's what I see in my email headers this morning:

X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on www550.hostpc.comThis of course is an upgrade to what we had before the server move:

X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on www534.hostpc.comThanks guys!

From this thread: http://hostpc.com/community/showthread.php?4316-EAS-being-disabled-on-all-servers

It took us some time to figure out - the installer kept throwing an error - we found ONE simple little perl module that needed to be updated, hidden in about 10,000 lines of error code. Once that was found, bam - done.

Sorry for the delay in Spamassassin folks, we literally were working full bore on getting it running.

[esgdirect]

Thank you! No Spam today and time for my index finger and left mouse button to heal. Lets hope all is well for a long time now.

[admin]

Thank you! No Spam today and time for my index finger and left mouse button to heal.

lol - I hear ya.


Joe

[Domaschuk]

Now that SA has been enabled (although I am still getting quite a bit of spam coming through) when will the other services like Installatron be re-enabled?

Thanks

[admin]

There's currently no known problem with Installatron on www550

[Domaschuk]

There's currently no known problem with Installatron on www550

I am actually on 533, but I entered a ticket a couple of days ago and it was summarily closed with no comment other than a link to this thread, so obviously I assumed that the two are related.

Anyway, after a week or so of trying to access it - including just about two hours ago - it still wasn't working. However, I tried again 5 minutes ago and now it is up.

[atene]

Since we no longer have EAS will catchalls again be offered as still advertised on your DA link from your host package page.

are we all still on our own to find out how to email people on ATT, SBC and AOL. This certainly seems like a host issue.

[admin]

Since we no longer have EAS will catchalls again be offered as still advertised on your DA link from your host package page.

are we all still on our own to find out how to email people on ATT, SBC and AOL. This certainly seems like a host issue.

It's definitely a host issue. I personally put a request in (actually multiple) with ATT/Comcast/Verizon and AOL (they handle a lot of the smaller filtering also) - a MONTH ago. I've hounded them weekly, and still no action.

I'll be on the phone with them again tomorrow.

Joe

[admin]

I just resubmitted to all 4 again - got the same responses:


We have received your AOL Postmaster support request. Currently, we are reviewing your request.

[admin]

Now that the www550 fiasco is behind us, it's time for an update.

First, the reasons for the www520 and www534 HUGE delays. The write access on the drive array was significantly failing. It was taking disks in the array multiple full seconds to sync - when it should be milliseconds. That was causing huge issues, as many of you know. Backups were no exception, but I, personally, underestimated HOW bad it was. (We honestly couldn't even get a good daily backup image of these servers every night). Now that the customers from those two servers have been migrated, to a correctly configured, high speed server with a significant amount of resources behind it (we spent 3x the price for this server), things, as far as I can see, have improved DRAMATICALLY.

With that being said, I've arranged for 3 more similar servers, with the same resources, to begin migrating additional customers away from the already loaded servers. Enabling SpamAssassin created an additional load, admittedly, but the new servers are handling loads very well.

We have already had several customers move from our next most populated server, a reseller box (www609) to a newer server. That's helped the load tremendously on that box. I'd like to move a couple more clients off that box ideally, and we'll continue to work with the resellers on it to make that happen.

Another cluster of servers of concern is www522, www530 and www531. I'd like to begin migrating those soon, though no time table has been established yet. To address the issue of emails not being sent (and I confirmed they did never go out, but I don't know why, it's certainly not my style) - I will send no less than TWO emails to all customers on each server being changed. They will be sent to the address in your Directadmin control panel. Like the first two servers, the master server IP will change. The IP for those servers has already been white-listed with the major RBL's (unlike www550). DNS will not need to be changed, with the exception of those just pointing to the IP with custom DNS services. We'll update that information in the emails being sent out.

I'm running test runs of the backups to insure they're NOT going to take anywhere near as long as the first cluster**ck. I won't proceed until I'm confident we can get it done quickly and with the least inconvenience to the users.

Lastly, I'm aware that there's been many questions, speculations and few answers as to HostPC. During August, several events happened for me personally, including some serious medical issues and a trip to a family wedding out of state, which took much longer than I expected - and included Hurricane Earl. I'm recovering, slowly - but am recovering. (thank you to all who sent get well wishes). As for the rest of the questions, unfortunately, I can't address those questions at this time (I promise, it's nothing "bad") - but you have my word that I'm working to bring HostPC services and commitment to our customers back to the level you expect. I ask that you have patience (again) with me while I work to make this happen.

Thank you again - I appreciate your support. If you have any questions on the upcoming moves, please let me know.

Joe


PS: If you're having issues with forum registrations, we're hammered with spam registrations - it's really a good idea to register here, then open a helpdesk ticket with the username or email address you used, and we'll get it approved. I hate spammers :(

[taylor]

Thanks for the update Joe. I appreciate your commitment to getting HostPC back to the level many of us have come to enjoy and expect. I've been holding off on this question until the dust from the last move settles a bit. Here it is:

Does running suPHP make it possible for some directories that currently seem to need 777 permissions to be able to have 755 permissions? Let me explain... I currently have an IPB forum on www536 that requires that certain directories have 777 permissions. I believe this is because writes to the directory are written with "apache" as the user/owner. If the permissions are set to 755 it doesn't let the "apache" write. I assume that suPHP allows those writes to occur with the user of the account rather than the apache user, but I haven't really read up on suPHP, so I'm just guessing here. In a related question: Is it possible to configure things pre-server move so that they will work correctly post-server move, or do I need to wait until after I my account has moved in order to make changes?

TIA

[Andrew Elmore]

for those missing installatron - look to the right side of the menu at the bottom
at first I thought it wasn't there either but it had just changed its place in the menu.

the move was painful but everything came out ok for me in the end.
I'd note I use zoneedit for DNS and Google for email hosting and have set zoneedit up for a fail-over redirection (not a duplicate but something so it doesn't look like we've disappeared).

[admin]

Thanks for the update Joe. I appreciate your commitment to getting HostPC back to the level many of us have come to enjoy and expect. I've been holding off on this question until the dust from the last move settles a bit. Here it is:

Does running suPHP make it possible for some directories that currently seem to need 777 permissions to be able to have 755 permissions? Let me explain... I currently have an IPB forum on www536 that requires that certain directories have 777 permissions. I believe this is because writes to the directory are written with "apache" as the user/owner. If the permissions are set to 755 it doesn't let the "apache" write. I assume that suPHP allows those writes to occur with the user of the account rather than the apache user, but I haven't really read up on suPHP, so I'm just guessing here. In a related question: Is it possible to configure things pre-server move so that they will work correctly post-server move, or do I need to wait until after I my account has moved in order to make changes?

TIA

777 will not work with suPHP - 755 will. No, you can't configure ahead of time, but during the transition, we WILL recursively set everything to 755 for you, along with the ownership changes.

[admin]

Not very stable NS531 if you ask me...Every few hours something resist and wipes out all users who have logged into my forum and force then to re establish a login to the site so how was this better then before???


If your users aren't setting to cookie correctly (software controlled), that's about the only reason they'd need to re-login. That's not normal, but sounds specific to the software you're using.

As I said, www531 is on the priority list of the next batch to change

[admin]

Regarding www550:


Dear Administrator:

Thank you for contacting the AT&T Postmaster.

The mail server IP address(es) associated with your request will be
removed from the block list within 24-48 hours from the date of
this notification. AT&T, Bellsouth, SBC and any affiliates do NOT
intentionally block legitimate mail in the course of our anti-spam
initiatives and regret any inconvenience this may have caused. If
the IP that was recently blocked begins to exhibit the
characteristics of a compromised network object or is compromised
by an offender of Acceptable Use Policies, the IP address will be
blocked again.

Thank you for helping the AT&T WorldNet Services network combat
spam in all its forms.

Regards,

AT&T Postmaster

[acm]

Now that the www550 fiasco is behind us, ...

Does that mean everything is supposed to be complete? I ask that because there are still a couple things not back to the way they were before the move:

1) I still can't access email via port 995 (SSL/TLS), something I always used prior to the move. It appeared to work for a short period of time yesterday morning but since then it's back giving "Connection to server timed out." errors.

2) I have SSL enabled on one of my sites and prior to the move I could control the SSL certificate but now when I try that I get an error saying "Cannot Execute Your Request, ... You can only add a certificate if you own the ip you are using".

I haven't opened a ticket yet because I assumed the two were related and I when I mentioned the 995 port issue on this forum earlier last week and the response was it was on the todo list.

[admin]

Does that mean everything is supposed to be complete? I ask that because there are still a couple things not back to the way they were before the move:

1) I still can't access email via port 995 (SSL/TLS), something I always used prior to the move. It appeared to work for a short period of time yesterday morning but since then it's back giving "Connection to server timed out." errors.

2) I have SSL enabled on one of my sites and prior to the move I could control the SSL certificate but now when I try that I get an error saying "Cannot Execute Your Request, ... You can only add a certificate if you own the ip you are using".

I haven't opened a ticket yet because I assumed the two were related and I when I mentioned the 995 port issue on this forum earlier last week and the response was it was on the todo list.

SSL can only be completed with a SSL certificate, which requires a static IP address that's available from the support center ordering process.

[acm]

SSL can only be completed with a SSL certificate, which requires a static IP address that's available from the support center ordering process.

I've had a static IP address ever since I've been with HostPC in January 2008.

[admin]

Then I'd suggest opening a ticket, with your static IP address listed.

Thanks

[Ziptar]

Are there any plans for www532? I've been getting complaints about slowness and IMAP disconnects in web mail all week.

Thanks,

[acm]

Then I'd suggest opening a ticket, with your static IP address listed.

Thanks

Ok, I just opened ticket JAX-421556 for this purpose.

[tonydi]

Are there any plans for www532? I've been getting complaints about slowness and IMAP disconnects in web mail all week.

Thanks,
I think that pretty much all of the servers are overloaded to one extent or the other. I see the same sorts of things on www533. Things weren't all that snappy before EAS got pulled, but they can be positively glacial at times now.

[admin]

I think that pretty much all of the servers are overloaded to one extent or the other. I see the same sorts of things on www533. Things weren't all that snappy before EAS got pulled, but they can be positively glacial at times now.

load on www533 is below 1.0 - which is VERY acceptable. Before SA was enabled, the load was .8 or so, so it made no appreciable difference in this particular server. However, as I said, all servers will be addressed in order of severity

[admin]

Are there any plans for www532? I've been getting complaints about slowness and IMAP disconnects in web mail all week.

Thanks,

I haven't seen this issue occur yet - but yes, all servers will be addressed in order of severity.

[tonydi]

load on www533 is below 1.0 - which is VERY acceptable. Before SA was enabled, the load was .8 or so, so it made no appreciable difference in this particular server. However, as I said, all servers will be addressed in order of severity

1.0 would be heaven. I know it has been waaay higher than that at times since SA was re-enabled.

[ricej]

Regarding www550:

That's one down. Anything from AOL or SBCGLOBAL?

[admin]

That's one down. Anything from AOL or SBCGLOBAL?

Another request / phone call today, still no reply.

I've got ticket #'s for each submission.

It'd really help the world if they didn't, by default, block entire class c's by DEFAULT and only open (or whitelist) by specific request

[tonydi]

Good News / Bad News.....

Good news is that emails from www550 now go through to AT&T (sbcglobal.net) and AOL addresses.

BAD NEWS is that they both end up in their SPAM FOLDERS!

I'm not sure that's progress.

[MrK]

1) I still can't access email via port 995 (SSL/TLS), something I always used prior to the move. It appeared to work for a short period of time yesterday morning but since then it's back giving "Connection to server timed out." errors.


Actually, port 995 is for POP3/SSL access. You should be able to access POP3/TLS over port 110. I've never been able to access HostPC email over POP3/SSL 995 or IMAP/SSL 993, but TLS has worked for me on both the old and new servers.

[acm]

Actually, port 995 is for POP3/SSL access. You should be able to access POP3/TLS over port 110. I've never been able to access HostPC email over POP3/SSL 995 or IMAP/SSL 993, but TLS has worked for me on both the old and new servers.

I did use port 110 temporarily until they got my static IP straightened out. I'm back on 995 now and all is well.

[High Technology]

Joe,

It's now been 3 weeks since the infamous www.550 incident, and while things seem relatively back to normal, it seems that HostPC customers suffered immense inconvenience in this cluster**** (as it was referred to twice earlier).

I share the immense frustration of many others here -- and I (like presumably others) were unable to post in this thread since our forum sign-ups kept getting deleted.

For the record, I've been here for over 5 years, and have been an avid supporter of HostPC in other forums. I am loyal because I've always thought you provided excellent service and stood up for your customers. After such great service for many years, I for one became extraordinarily irritated when things when awry.

It all started when EAS was summarily turned off server-wide with NO notice. For those of us who've been around for a few years, we remember when SA just couldn't keep up. Some of us, myself included, paid extra for EAS initially, and really appreciated the value. I've seen more ads for fake little blue pills in these few weeks, I starting to question my own virility. :rolleyes:

- Kidding aside, please explain why was EAS turned off, and why was it done without any notice and without at least turning SA back on.

Regarding the server update, like others, I've been through several server updates over the years. Most of these were done with ample notice, and done overnight with little downtime. This time I got an email about an hour before my site went down that pointed to this thread, which is not adequate notice. If you are taking sites down, ample notice should be given to customers.

- We've already recounted all of the issues of the unexpected downtime and backup/restore issues in this thread so I won't rehash them here, but why wasn't there a "dry run" of the backup done as a test? I am not an IT expert, but it seems odd to try to backup/replace/restore a live server without having regression tested everything in advance.

The inconvenience caused by the downtime still lingers -- we were down long enough that some subscription based emails deactived the address since the server wasn't responding for a couple of days. I think I've now caught them all, but it's a little difficult to figure out what non-recurring emails might not have come through and may still be turned off.

With respect to the email issues, I appreciate that you asked for requested whitelisting the new ip addresses in advance, but again this is something that should have been tested before going live with two whole servers full of sites.

Out of frustration from all of the spam, and feeling like little was getting done on the server side, I opted to use Google Apps for my email engine now. That being said, it's SPAM filtering seems to be almost as good as EAS, but it did flag some very legit sites (Staples being one) -- I've since flagged them as "Not SPAM", but I am a tiny bit leery that some legit emails might get flagged. But it's WAY BETTER THAN SA. I recall you making a reference for using Google above, but it wasn't clear if you were talking about plain vanilla GMAIL or Google Apps with your own domain name.

- I wish I kept a step-by-step set of screen shots so I could have documented what I did as a tutorial. Unfortunately, as my first attempt at this, I was unnecessarily hesitant with all of Google's disclaimers. All-in-all, it was actually a pretty easy process. Perhaps if this is a longer term solution to keeping the server loads in-check, perhaps one of your support staff could document the steps for others to follow. This seems like a win-win situation for both the users and HostPC.

Lastly, since it's been several weeks, perhaps you might want to confirm or dispel the rampant rumours that HostPC has been sold, and that there are new owners. A change in ownership / strategic alliance / etc. is often a good thing and not a bad thing, but you've just ducked all of those questions suggesting more info will come soon.

While I can't speak for others here, I can see that you have a loyal customer base, one that understands that we are all human. With that said, I for one would like to hear the answers to these questions and understand who/what HostPC is now.

Thanks for hearing me out and for considering my questions.

[taylor]

Thanks for posting this High Technology. I'm wondering many of the same things, and I appreciate the care with which you have phrased the questions.

I've been toying with switching to Google Apps for email and was interested to see your assessment of its spam filtering being superior to SA. I'm curious, are you making use of a catch-all address there? I have too many aliases (300-400) to set them up in Google Apps so I'd have to use a catch-all. I'm concerned that it would result in more spam getting through.

[tonydi]

Nicely written, High Technology, and as someone who has been hear for over 6 yrs, I share your viewpoints on all that has gone down. But as I've said in other threads, no answers will be forthcoming, at least not from Joe. He personally told me that he will not and can not answer any of the questions we all have. Because of this and other thing's he's said, it's clear to me that Joe is no longer at the helm and was only brought back when the current "tech" support people (person?) couldn't deal with the server problems, RAID failures and subsequent re-enabling of SA after the server move. Moving to Gmail filtering only solves one of the problems here. As has been said numerous times since this all blew up, back up your sites frequently!

[Jerry]

You know, that's an interesting point. If Joe is no longer the owner, shouldn't that be visible on some public record/search?

Keeping the owner hidden would seem to be quite questionable at best. If, and that's a big 'if'... If the new owners of a company didn't openly inform its customer base, but rather had a need to deceive the customer base, then I'd say they are playing a bad game Russian Roulette. It would likely come back to bite them one day.

Simply put, if they have a need to keep information from the customers, and they deceptively act on that rather than be up front and honest, then that act is the very reason why we should question if we are going to give the new ownership our trust. I have faith in Joe and crew to be able to support my hosting needs, but would I trust a new owner? Well, I'd give them the benefit of the doubt. At least as long as they were up front and honest about what was happening.

I don't know that I'd buy into gossip about new ownership though. We should be careful not to start rumors and such, needlessly.

[High Technology]

Thanks for posting this High Technology. I'm wondering many of the same things, and I appreciate the care with which you have phrased the questions.

I've been toying with switching to Google Apps for email and was interested to see your assessment of its spam filtering being superior to SA. I'm curious, are you making use of a catch-all address there? I have too many aliases (300-400) to set them up in Google Apps so I'd have to use a catch-all. I'm concerned that it would result in more spam getting through.

Unfortunately, I don't use a catch-all account, so I don't know if that would reduce the effectiveness of the SPAM filtering.

The set up is pretty easy and you can quickly and easily change the MX records back if things aren't working out with Goggle Apps.

You can also try Google Apps Pro free for 30 days, and that would allow you to get dual delivery, perhaps using a few of your addresses as a test. If it works, downgrade back to basic and use it.

[taylor]

You can also try Google Apps Pro free for 30 days, and that would allow you to get dual delivery, perhaps using a few of your addresses as a test. If it works, downgrade back to basic and use it.
Great idea. I'll try that when I have some time to mess with it.

[mharvey]

I use a catch-all with Google Apps on my domain and I have not noticed an increase in spam. Google seems to handle it very well. Google Apps does support what they call "nicknames" that can be used to create aliases for email accounts that you have created. I can see that it would be time consuming to set up that many but I think you could do it if the catch-all did not work.

[taylor]

Thanks. I'm familiar with the nicknames, but they only allow 30 per email account.

[ctechnow]

Are there any updates here? I'm in the process of migrating to www611 and cannot email AOL or SBCGLOBAL.

[hockey-dad]

similar to many stores my website makes most of its money between the Sunday before Thanksgiving and New Years. PLEASE Joe, no server upgrades during the holidays. :love

[mharvey]

1.0 would be heaven. I know it has been waaay higher than that at times since SA was re-enabled.

I know that I have seen loads MUCH higher than 1.0 on www533 recently. As a matter of fact it is showing at well above 10.0 at the moment:


www533.hostpc.com DISK USAGE: UPTIME: 10:28:16 up 6 days, 1:41, 0 users, load average: 14.26, 9.53, 7.71

[tnas]

Where do you get these server numbers from? Do you have to have a reseller account to see them? I don't see them on my single user account (at least I haven't found them).

[amelith]

tnas,

Your nameservers should tell you. For example, my nameservers are ns537a and ns537b, and I'm on www537.

[tonydi]

No, he's talking about the average load numbers. A normal server should be around 1, or 2 at most. Those numbers from www533 show a severely overloaded server, one that probably was giving timeout errors accessing web sites and email.

[amelith]

Oh, ha, sorry. My bad. Yeah, load numbers would be nice to have, though I too am on a user account.

[mharvey]

http://www533.hostpc.com/report.txt

Over the past several weeks I do not think I have seen www533 report a load under 4 or 5... The past 24 hours have been much different. I also note that over that period of time I could never get the Direct Admin file manager to open up without a time out. Today... it works just fine. I wonder if anything has changed:


www533.hostpc.com DISK USAGE: UPTIME: 20:14:01 up 7 days, 11:27, 0 users, load average: 1.81, 1.51, 1.37

A server load of 1.8 is still high but not even near the horrid loads I was seeing earlier.

[tnas]

Thanks for the link mharvey
www537 shows:

www537.hostpc.com DISK USAGE: 28% /dev/simfs UPTIME: 08:53:01 up 47 days, 2:11, 0 users, load average: 15.11, 15.79, 13.13

</pre>