softlayer server auto reboot ?

Portal Home > Knowledgebase > Articles Database > softlayer server auto reboot ?

Posted by DaweiRui, 04-13-2007, 03:32 PM
I'm use softlayer server 24 days , it seems all ok , fast and cool . But just now , in my watch , the server can't ping , after 10 minutes it's back . I'm check the log messages , it happened in Apr 13 14:08:28 . I don't know it's AUTO reboot ? or softlayer doing something ? or my new server large Load Averages ? server justLoad Averages0.12 0.09 0.02 any softlayer user guys ? any advice ? thanks ...
Posted by Xandrios, 04-13-2007, 03:46 PM
Check what happened before the reboot? If it was manually rebooted it should say in the log from where the command was issued.
Posted by eth00, 04-13-2007, 03:47 PM
Softlayer apparently had some sort of power issue, nothing official yet but it was out long enough to just kill power and reboot the servers. There is also a thread on the softlayer forum on some user comments but at this point nothing official.
Posted by DaweiRui, 04-13-2007, 03:53 PM
There is also a thread on the softlayer forum on some user comments about some sort of power issue ?? that's meaning not my server problem , right ? thanks , Xandrios and eth00 reply .
Posted by steven-v, 04-13-2007, 04:05 PM
Huh ? ALL my servers was rebooted 54 minutes ago - right now uptime on ALL servers like 54 min. Must be some power-related issues thats affect whole SL.
Posted by eth00, 04-13-2007, 04:12 PM
[QUOTE=DaweiRui;4459545] that's meaning not my server problem , right ? /QUOTE] Correct not your server. We don't have an official answer but something is going on.
Posted by SoFiMaN, 04-13-2007, 05:55 PM
Yea, same happened here on one of our servers at SL but nothing happened to our other servers there, which means that not all of SL servers faced this downtime.
Posted by geedeedee, 04-13-2007, 06:08 PM
I am cancelling my servers with SL right now because of this.
Posted by futurehosting, 04-13-2007, 06:28 PM
Why? For a power outage which was resolved in under four minutes? Equipment can fail at times. It's the nature of the beast. Technology is bound to fail. If this has scared you off, good luck with most of the other providers. I've seen other datacenters down for hours while SL is almost always up and if there is an issue, is resolved promptly.
Posted by steven-v, 04-13-2007, 08:18 PM
How old are you ? 14 ? Here is much worst outages been in ALL datacenters...
Posted by Ukyo, 04-13-2007, 08:47 PM
The Infomart in Dallas (Where SL is) just took a small power hit. However it was only out for 42 minutes (per our UPS). I dont know if SL's outage is the same one or not, but it seems like its possible. Update: Actually, looking at the timestamps of the forums, it looks like they may not be related. Last edited by Ukyo; 04-13-2007 at 09:01 PM.
Posted by vengefuldeath87, 04-13-2007, 08:51 PM
the softlayer downtime was due to a switch failure on one of the ups units during a routine maintence.
Posted by digdugdog, 04-14-2007, 12:03 PM
I guess no one noticed the weather in the Dallas area. I'm just glad the datacenter didn't get ripped apart! http://www.chron.com/disp/story.mpl/...n/4714235.html Only some of my servers were affected, one disk went into read only mode. Total softlayer downtime for me was 4 hours for one server.
Posted by tsj5j, 04-14-2007, 01:09 PM
To be fair to SL, technology failures do exist, and are not uncommon. They do have a backup UPS, but if it fails they really can't do much, except have more backups and increase costs for the end user, which ultimately might all fail at the same time (yea, it happens).
Posted by eth00, 04-14-2007, 03:59 PM
To Softlayer's defense the outage was only about 1-2 minutes for most people plus the time it took to boot back up. Sounds like that server was running a FSCK because of the reboot.
Posted by HD Fanatic, 04-14-2007, 09:23 PM
hope softlayer resolves these problems and won't happen again, I was looking into getting a server from them.
Posted by futurehosting, 04-14-2007, 09:29 PM
There aren't ongoing problems. A bypass switch failed during scheduled maintenance. Hardware isn't always going to work. It is still a piece of equipment and will break eventually which is why there is redundancy. In the case of this, if the switch failed, there is nothing anybody could do. Was it unfortunate? Of course. But, at least the infrastructure was back up quickly. Our servers were rebooted and back online within 4 minutes. Some of the other SL customers had to go through a FSCK. It's part of life and must be dealt with. At least power wasn't offline for hours like some other places we have been with.