I am wondering why Geoffrin is so unstable. For how many hours was it down?
how about the Opteron's temperatures? I suppose it is installed in an air-conditioned room.
and how about the Opteron's motherboard BIOS? is it updated with a stable revision? Some Opteron motherboards had problems with the BIOS.
for the remote reset issue somebody suggested this on http://openfacts.berlios.de/index-en.phtml?title=Wikipedia_Status :
"Type "remote power management" into Google, and you'll get listings for lots of network-controlled power-distribution devices, which will allow you to remotely power-cycle locked-up boxes, but without the cost of a UPS, and more importantly, with the ability to power cycle on a per-power socket basis. (wikipedian, the Anome)"
-- Optim
__________________________________ Do you Yahoo!? New Yahoo! Photos - easier uploading and sharing. http://photos.yahoo.com/
On Dec 26, 2003, at 16:44, Nikos-Optim wrote:
I am wondering why Geoffrin is so unstable. For how many hours was it down?
Once it crashes it stays down until someone reboots it. The length of the outage isn't related to the seriousness of the cause.
how about the Opteron's temperatures? I suppose it is installed in an air-conditioned room.
Who knows? How to check?
and how about the Opteron's motherboard BIOS? is it updated with a stable revision? Some Opteron motherboards had problems with the BIOS.
Who knows? Who can check?
for the remote reset issue somebody suggested this on http://openfacts.berlios.de/index-en.phtml?title=Wikipedia_Status
We've *got* a remote-controllable power strip. We're out of outlets, and it doesn't allow cycling two outlets at once.
-- brion vibber (brion @ pobox.com)
Brion Vibber wrote:
how about the Opteron's temperatures? I suppose it is installed in an air-conditioned room.
Who knows? How to check?
I don't know how to check the internal temp, if anyone knows, that'd be cool. But I can say that the facility that it's in is perfectly fine -- a proper modern colocation facility with ample ventilation, backup power, and what not.
We've *got* a remote-controllable power strip. We're out of outlets, and it doesn't allow cycling two outlets at once.
I'd be happy to move some less urgent machine of mine off the remote power strip, to make room for wikipedia, except that the limiation on cycling two outlets at once makes this a useless gesture.
But, remote power strips are not expensive in the grand scheme of things, if someone can find one that permits us to cycle two outlets at once. This can't be a rare problem -- tons of people have machines with redundant power supplies.
(My bright idea was to leave one of the outlets unplugged for now, so we could cycle the machine by cycling the other, but Jason reports that the machine makes an ungodly noise when both supplies aren't plugged in!)
I'll be on the phone to Penguin Monday morning, consulting with them about a sensible course of action.
--Jimbo
Jimmy Wales schrieb:
how about the Opteron's temperatures? I suppose it is installed in an air-conditioned room.
Who knows? How to check?
What we need is called lm-sensors, a kernel module to read out internal motherboard sensors. But I've never done that on Linux. My understanding is, that you also need special drivers for i2c bus and whatnot. Someone with the specs of Geoffrin at hand could figure it out.
If you are lucky, the SuSE kernel we are using now has all this stuff precompiled, all you'd have to do is to figure out what modules must be loaded.
Once you've got the lm-sensors driver working, it's easy to find a userspace frontend for it that can print out the current temperature of the processors on the console.
Alwin Meschede
the temperature check should be performed while the CPU is working at full load (100%). to raise the CPU usage at 100% you can use some benchmark software like www.mersenne.org (run the software twice, for both CPUs)
you can do the check with: 1. software, as ameschede@gmx.de wrote. 2. BIOS, if it reports temperatures. in this case you need a console (keyboard, monitor) and physical access to geoffrin. Warning: the temperature you read in the BIOS is NOT the temperature of the CPU when it works at full load (100%). 3. install a thermal sensor. not a very good idea. 4. open the server's box and place your fingers on the CPU :)
also check the motherboard manufacturer's website and see whether your BIOS version has some bugs. in that case, install a newer BIOS, but NOT the newest one (to avoid new problems, newest BIOSes are untested).
Optim
--- Alwin Meschede ameschede@gmx.de wrote:
Jimmy Wales schrieb:
how about the Opteron's temperatures? I suppose
it is
installed in an air-conditioned room.
Who knows? How to check?
What we need is called lm-sensors, a kernel module to read out internal motherboard sensors. But I've never done that on Linux. My understanding is, that you also need special drivers for i2c bus and whatnot. Someone with the specs of Geoffrin at hand could figure it out.
If you are lucky, the SuSE kernel we are using now has all this stuff precompiled, all you'd have to do is to figure out what modules must be loaded.
Once you've got the lm-sensors driver working, it's easy to find a userspace frontend for it that can print out the current temperature of the processors on the console.
Alwin Meschede
Wikitech-l mailing list Wikitech-l@Wikipedia.org
http://mail.wikipedia.org/mailman/listinfo/wikitech-l
__________________________________ Do you Yahoo!? New Yahoo! Photos - easier uploading and sharing. http://photos.yahoo.com/
wikitech-l@lists.wikimedia.org