OVH Community, your new community space.

I am sick of complaining but everyday something else...


Thelen
24-08-2010, 12:00
The problem is the HDD getting too hot, not the CPU (for obvious reasons).

I've already tested everything but the failure isn't 100%, for all I know it isn't the HDD but that is the only thing that is getting remotely hot. Could be the mobo north/southbridge (which isn't measured by the sensors).

If you really wanna follow it up, have a look at 209075 and 209077 both in the same rack and ordered at the same time.

Heh, over half of the tickets auto created are from those 2 servers crashing XD

marks
24-08-2010, 10:32
Quote Originally Posted by Thelen
I did, nothing will/has/can be done. You simply run the DCs too hot, or especially this particular rack.
If the CPU is above the recommended working temperature, I can tell that it would be intervened. I've seen cases before, that can happen.

If that's the case, let us know. A ticket with the output from lm-sensors showing the high temperature should do. Let us know if you want us to follow it up.

Tz-OVH
24-08-2010, 09:52
Quote Originally Posted by Thelen
I did, nothing will/has/can be done. You simply run the DCs too hot, or especially this particular rack.
Not a bad idea for a thread? Lets ask users to post their CPU/MB/HDD temps

Thelen
24-08-2010, 05:49
Quote Originally Posted by marks
If there is a temperature problem, it'll be sorted. That's not very common but it happens. So, show it to us.

@Jonathan: that was a very good list of suggestions. This kind of investigation is exactly what it's expected from the server administrator.
I did, nothing will/has/can be done. You simply run the DCs too hot, or especially this particular rack.

marks
23-08-2010, 16:26
On your comments on the availability on the phone: it's true, this morning we had lots of calls and as usual, plenty of people on holidays.

Sad but true, specially for the ones left behind

marks
23-08-2010, 15:54
Quote Originally Posted by Thelen
I have two servers fail once a week due to heat, I've never bothered to complain because I know there is nothing they can do. They won't just turn up the AC. When I did complain via a side issue, they said update BIOS will fix, so I let them do it, surprise surprise still crashes once a week..
If there is a temperature problem, it'll be sorted. That's not very common but it happens. So, show it to us.

@Jonathan: that was a very good list of suggestions. This kind of investigation is exactly what it's expected from the server administrator.

olliegooch
23-08-2010, 12:32
Quote Originally Posted by Thelen
Lulz @ thread. Used to think I was the only one that was feeling as if OVH was a month long hangover.. now seems I'm not alone.

Can't wait to move to LW fully eh?

Also, when are servers arriving, 4 people have asked in a certain thread when they are coming, its been like 2 weeks past the 1 week for the interface thing you mentioned before.
LW?

Thelen
23-08-2010, 12:12
I have two servers fail once a week due to heat, I've never bothered to complain because I know there is nothing they can do. They won't just turn up the AC. When I did complain via a side issue, they said update BIOS will fix, so I let them do it, surprise surprise still crashes once a week..

I'd think he has tried all your suggestions, he does,after all, manage somewhere around 500 servers for his clients...

yonatan
23-08-2010, 04:11
Quote Originally Posted by RapidSpeeds
It's down again, it was alive for 2hours.

I've been trying to call Support for the past hour, what message do I get this time? All Advisors are busy, call back later.

I am disgusted by this to say the least...

10pm: Still can't get through to support & server still down.
hey man,
if the servers was down came back up , and went down again after two hours, i would suggest rescue-pro and start digging


before contacting support i always run this checklist: ( if one of them fails contact support or fix manually according to the issue )

does it boot in rescue?
what is the temperature of the server? ( too high in logs? contact support )
what is the status of the mdadm? ( broken? fix manually ).
keep in mind! if the server was hard rebooted, it might go into auto fsck check which might take hours to complete on some servers ( depends on md status+number of disks ).

did it pass hw tests? ( if mdadm failed to rebuild - check manually rebuild ).
nothing works? MemTest ... fail? contact support
CPU test.? fail?.. by now this is a rare situation only seen 1 CPU fry , contact support about this.


have all of these tests passed?

if it goes up and then down... something is off, either with a hardware component which you need to diagnose , or a software fault , which you need to diagnose.

check the cooling first of all... that's the most common issue.

Thelen
23-08-2010, 02:28
Lulz @ thread. Used to think I was the only one that was feeling as if OVH was a month long hangover.. now seems I'm not alone.

Can't wait to move to LW fully eh?

Also, when are servers arriving, 4 people have asked in a certain thread when they are coming, its been like 2 weeks past the 1 week for the interface thing you mentioned before.

turbanator
23-08-2010, 00:10
i feel ya bro..people are switching to other options for dedicated servers, OVH has gone downhill forever and now there is a barrier that they crossed that even people who wanted to save money are fed up.

SLA

If your server becomes unavailable, Ovh guarantees intervention and repair time on levels 1 and 2, from 30 minutes to 4 hours, 24/7. The availability of the network is 99.9%, 99.95% and 99.99% monthly. In case of non-compliance with the SLA, penalties are automatically calculated.

taken from the ovh website ^

RapidSpeeds
22-08-2010, 21:37
It's down again, it was alive for 2hours.

I've been trying to call Support for the past hour, what message do I get this time? All Advisors are busy, call back later.

I am disgusted by this to say the least...

10pm: Still can't get through to support & server still down.

RapidSpeeds
22-08-2010, 18:11
Quote Originally Posted by Myatu
Is that the UK 24/7 number you've called? (020 3384 2356)
Of course.

The server is back up now thankfully, but that was about 4hours with no english speaking staff.

Myatu
22-08-2010, 16:58
Is that the UK 24/7 number you've called? (020 3384 2356)

RapidSpeeds
22-08-2010, 16:54
So i've had a supposed SLA server down for 5hours now.

No reply to the automatic intervention ticket (which was only raised when I hard rebooted for the 5th time as the server was non-responsive).

So, I call the incident team and I am told call back in 30mins, system failure.

When I call back, I am told there is no English speakers available, call back in 30mins - I called back, then I was there was still no one available, try again in an hour - I take into account it's a Sunday, and as much as I find it laughable a company as big as this does this to clients, I need this server online.

Can I call UK Support and get some proper help? No I can't because they think servers only break down between Monday-Friday 9-6.

To sum my experience up with in a few words... OVH, you are a ****ing joke.