Mandrake Linux Archives: cooker-amd64@linux-mandrake.com
Mandrake Linux: cooker-amd64@linux-mandrake.com
[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
- From: Duncan
- Subject: [cooker-amd64] Re: Re[3]: SATA problem
- Date: 26 Jan 2004 13:40:17 -0000
Rui Pimenta posted <200401251943.06100.rpimenta@runbox.com>, excerpted below, on Sun, 25 Jan 2004 19:43:05 +0100: > One question for Duncan, if possible. About the opteron CPU. Mine is the > 242 and on its specs is reported that maximum case temperature is 69 > degrees celsius. I installed it 3 days ago and the normal temperature > reported on the BIOS is 52 degrees and the room temperature is around 23. > My case is a Thermaltake Xaser III with 8 fans. My doubt is if this > temperatures is a bit high or not, because I am afraid that I haven't > placed the heat sink properly. Mine is a 242 as well. However, I haven't gotten lm-sensors working, so can't check it under operational conditions, and don't reboot that often, to know what it's running. If with that question you are implying the stability isn't what you expected, and are wondering if you are overheating the CPU, I had stability issues myself, earlier, but they seem to have straightened up, now. I don't know for sure what fixed it, but if you are running a kernel older than 2.4.22-27mdk, that might be it, as my stability DID improve after loading it (I was running the -24mdk that came on my ISOs, b4 that). I'm just not sure if there were other factors as well, or not. Other things that may have played a part.. a) I decided to de-complicate my life a bit and uninstall shorewall and IPTables, as I have a NAPT based gateway-router between me and the internet, with no open ports and additional filtering. That could easily have been a destabilizer. b) I BELIEVE I'm running more stable now, after recompiling my own kernel, still from the -27mdk sources, but with reiserfs, my operational fs of choice, compiled in, to avoid an initrd, with most of the modules turned off as unneeded (which shouldn't affect stability if they weren't loaded anyway, and why would they be loading if I didn't need them, but..), with grsecurity turned on at low level (could affect stability, wasn't on by default), and with various other misc. changes. c) I had a number of lockups that seemed audio related (the amd8111 built-in sound, using the alsa-i810 driver), but had other lockups as well, so am not sure on that. However, I'm now thinking they may have been KDE-ARTS related, and have KDE set to NOT turn on ARTS, for sound effects, but play XMMS using the ALSA driver for music with little or now problem. Again, this might be the kernel update, or it might be ARTS related, or it might have not been sound related at all, but without ARTS running, I definitely seem to have fewer stability issues. d) This one shouldn't apply to you, but some of my initial problems were related to a Western Digital hard drive going out on me b4 its time. I normally buy Maxtor, but relented and bought WD for that drive due to getting a good price on it. However, that was a mistake. I've now replaced it with another Maxtor, which I've never had problems with. Anyway, at least PART of my initial sound problems were related to that hard drive, since when I was copying stuff over (I was able to ignore the SMART warning and spin the drive up, to retrieve most of my data, but gave up on some of it), the MP3s dirs were some of the worst affected, and I'm sure some of my initial freezes were simply I/O problems when XMMS tried to access the next MP3 or whatever, and it froze the system due to I/O issues. However, I still had issues after that with audio, at least thru ARTS, so the problems may have exacerbated each other, with the system crashes while attempting to access those MP3s probably not so healthy for the hard drive, either. While discussing hardware, memory can be a problem, but shouldn't be one here, due to the fact that I'm not overclocking it, and I've had a policy for several years now of purchasing ONLY ECC memory, which can correct small errors, including the single-bit errors typical of a bad chip, and should detect larger ones even when it can't correct them. e) It's of course possible some of it is related to additional updates as I've been continuing to update from amd64 cooker fairly regularly. However, in theory at least, regular applications aren't supposed to be able to bring down the kernel and the entire system, only themselves. We all know that theory doesn't always equal reality, however. Anyway, whatever it was, I'm running much more stable, now, but it took me a couple months to get there. Maybe it was just a system burn-in issue or something? So.. that gives you something to work on if it is indeed stability issues you are seeing. BTW, if you have lm_sensors or whatever running there, perhaps you could give me some pointers on getting that running here? I never ran it on i586 either, so don't know for sure where to start in trying to figure out why it doesn't seem to work for me here and now, tho I know both the sensor chip, and the i/o buss, are supported. (lm85, and amd8111, the modules do load, but sensors says no sensors found). The README says the 2.4 kernels need different modules compiled, but I'm not sure if that applies to the Mdk distrib kernels or just the vanilla kernels? -- Duncan - List replies preferred. No HTML msgs. "They that can give up essential liberty to obtain a little temporary safety, deserve neither liberty nor safety." -- Benjamin Franklin
- References:
- [cooker-amd64] SATA problem
- From: Rui Pimenta
- Re: [cooker-amd64] Re: SATA problem
- From: david stephens
- [cooker-amd64] Re: Re: SATA problem
- From: Duncan
- [cooker-amd64] Re[3]: SATA problem
- From: Rui Pimenta
- [cooker-amd64] SATA problem
- Prev by Date: [cooker-amd64] suggest the right configuration
- Next by Date: [cooker-amd64] Re: suggest the right configuration
- Previous by thread: [cooker-amd64] Re[3]: SATA problem
- Next by thread: [cooker-amd64] Is there a right place to start?
- Index(es):
