New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
OVH Quality!
This discussion has been closed.
Comments
Well i really dont know, but all of them said its a hardware failure.
Are you serious?
I readed, and answered.
If your System Admins have said it's a hardware failure then provide OVH with the proof the System Admins gave you, if they gave no proof, they don't know for sure what it is.
Out of interest, are you running OVH Kernel or the Distributions Stock Kernel?
@AlbaHost
For how long have you been with OVH ?
Do you always order servers with AMD processors at OVH?
Any specific reason for using AMD processors?
I have provided OVH with sysadmins proofs
Yes, im running OVH Kernel.
I use OVH for almost 6 months ago, i had another lowe plan from hosting package and never had any error, till i get higher plan (current one)
Yes.
Not really.
That is where the issue cloud be. I always use native distribution kernels on servers where I setup virtualization stuff & never have had a problem.
Always use native distribution kernels & it might solve the issue for you. That is from 3+ years of virtualization experience on OVH servers.
Thank you, i will try to reinstall OS with a native kernel.
Sorry - must have missed that. I thought you said that your sysadmins found nothing which is why you suspected that it was hardware. (And you also said OVH replaced hardware even though there was nothing that actually tested wrong with it but the system continues to crash.) So what was the proof your sysadmins gave to OVH?
Something's not making sense. From your OP the load is insanely high. Without really delving into anything it would seem you are just driving the system to the ground until it locks up hard (at which point of course nothing would be written to logs).
Do you have any active monitoring so you see what is happening on the system right before it crashes? Something like munin would go a long way.... You could track what all of your different processes are doing (mysql, mail, http, etc) and also see how they correspond to system events (high memory usage, high i/o, etc) and break things down that way to really get a handle on whats happening on the system before it croaks.
As already suggested use the Distributions Standard Kernel, I've personally had issues in the past with the OVH Kernel causing crashes, I've heard of other people having the same issue as well.
You don't need to reinstall OS to use native kernel.
I have seen some issues in OVH kernels too.
What does OVH change in their kernels anyway?
On centos, selinux was permanently disabled. CPU likes to stay at low frequency even with scheduler set to permanence.
The problem is the inexperience on the ops behalf. Get out the hosting business!
Your disks are failing, good luck getting ovh to replace them though.
Quote of the day
Other than that, everyone who diagnosed hundreds or even thousands of hardware defects knows there are occasional hard ones. Heck, I remember once I swapped "everything" and there were still crashes. It took me some time to think change the sata cables too, and magically, the crashes stopped. Failures were so hard that no log was written, maybe the drives were just locked.
Another hard one is when some os/app crash on a certain hardware. I had issues like this with freebsd derivatives that just hanged.
But the most strange of all was on a production server on which could do only minimal downtime in the middle of the night and did that a couple of days a week for more than 3 weeks scratching my head, changing components, reinstalling tens of times. It was an AMD server and I did change the CPU too, to make sure, but XCP 1.6 refused to go on if there were mixed VMs inside (full and paravirtualized). Once there were only full, or paravirtualized, it worked without issues for days, start one of the other, it was crashing hard in 10-20 minutes. No logs, just reset. I closed the case as a hardware incompatibility, probably some issue with rings in the CPU and fully virtualized all VMs. At times the machine wins.
Yes, there are many similar cases. I once had a server in my office network (opteron, if I recall well and Windows 2003) and the machine hanged a lot due to memory failures. I changed memory , nothing, changed power supply, nothing, changed cables, nothing. I was trying to find a similar motherboard to test the rest hardware till a friend of mine suggested to change the... mouse! (mouse that was working like a charm in this server and in other pc's).
Voila, that was the problem (don't ask me how a mouse can create memory hang problems, I still don't know)... So, sometime the weirdest things can occur caused by hardware...