wolstech

Johnny isn't actually online. He's being rebuilt, and what you're seeing is a brand new, unconfigured install of cpanel, since we backed up the data and nuked the server yesterday. There are actually no accounts on Johnny right now. Once it's configured, krydos will probably restore all the user backups. Our account system still thinks all the Johnny accounts exist though, since they should be coming back concerned the server is rebuilt, which is why you can't reuse the name. Not sure why support@heliohost.org is bouncing for you. Doesn't matter in terms of actual support, since emailing that address actually just posts your email on the forum, but I'd guess the spam filters are being a little oversensitive. I'll check the log when I get a chance and see if the raw email even made it to us. Tommy isn't released yet, hence why he's always "full". It's currently an invite-only beta. Your best bet for now is to pick a different username and sign up in Stevie. If you had domains associated with your Johnny account you want to reuse, let me know and I'll have a root admin remove them for you.

The server basically froze. Your files should be there when it comes back up. The only thing I'd worry about is InnoDB database tables...those have a history of going corrupt after crashes.

You're not suspended, the server crashed. It's been rebooted and is currently running a disk check.

Thanks for the update Byron. Any idea on what happened to Charlie/Johnny in the meantime? He seems to have also died late last night.

Moved.

We currently do not have an estimate for when this will be restored. It's being actively worked on.

As Byron said, we don't keep backups. You have to make them yourself if you want them. In the same boat myself...I made a backup before an update I did over the holiday last week, then forgot to actually download the thing when it finished... I'm now missing that backup, plus all the work I did Thursday and Friday, and all the new data that got collected by my software on Monday and Tuesday. Oh well...I always preach "make backups!", and didn't follow my own advice.

Stevie is down.

how to move to tommy You can't unless you were invited since he's invitation only right now (he's not released). I just have an unused account on tommy since I'm an admin here...I finally have reason to put it to use. Wait for Stevie to come back up. Not sure what's going on with the second crash. Byron said krydos is aware of it.

...and down it goes again.

Yeah, cPanel seems to have broken in the crash earlier. The websites themselves are back up, but cP and NS1 are still not working correctly. This support request is being escalated to our root admin.

Websites should be back up. cPanel seem to still be down though (returns a 500 error for no such file or directory in the sessions directory), and NS1 is also still not back yet. EDIT: Looks like he crashed again. If he comes back up enough for me to get into cP, I think I'll grab a backup and consider moving to Tommy.

No, the server is down. See http://www.helionet.org/index/topic/26650-stevie-down/?view=getnewpost

It's already been rebooted as of a half hour ago: http://www.helionet.org/index/topic/26650-stevie-down/?view=getnewpost It takes quite some time for him to come back up though (disk checking and the like have to run).

Ok

Got 4 reds on the monitor and a pile of timeout alerts by email...

That domain is already configured as the main domain for that account. Is this the domain you want? It's showing queued, but they do take 24-48 hours to start working, so that's normal if you just changed it. Give it a day or so to finish setting up, then clear your cache. If it still doesn't work after that, let me know and we'll look into it further. EDIT: Byron beat me to it.

What account needs the domain changed?

Please clear your cache.

3 cron jobs. As you're aware, the limit is 2 per day. Unsuspended.

This support request is being escalated to our root admin.

Tommy is invitation only. Our public offering is currently only php 5.3.8. Please find an older version of your software.

I can't think of any time we lost files. InnoDB database tables are the most likely thing to get lost. InnoDB is horrible at fault-tolerance, its basically a 50/50 chance every time the database server shuts down improperly as to whether your InnoDB tables will still be there when it comes back.

Looks like the whole server crashed. I can't even think of the last time this happened. He sees load spikes and SQL issues rather regularly due to overloading, but it looks like this time the whole server fell over under the load.

Looks like he crashed...even the NS is down. Escalating. For the interested (maybe Krydos will better understand what I saw): Right before the crash I was on SSH editing some code of mine since I updated my main software today. I noticed HTTP slow then start timing out while testing. Top showed PostgreSQL and recquota (linux disk quota service) beating on the CPU (both pegging it). I left the SSH open with top running, and 10-15 minutes later the session died abruptly. Haven't reached Stevie since.