Run memtest from the Ubuntu boot CDs for a couple of hours, or even overnight.

I don't think pulseaudio does any kernel level work, so I doubt if it is a direct cause of a hang. It may be indirect if it was exercising something else that may cause a hang, such as some feature of ALSA. Describe the hang ... can the host be pinged on the network?

Okay, good, that excludes single cause memory or CPU problems. In my day job I've seen systems that will not trigger memtest failures yet show evidence of them when other devices are used. memtest doesn't exercise the memory at the same time as the PCI bus or devices on it. So it helps here by suggesting the memory and CPU are fine.

Thanks. Also, take note of what hangs ... does the mouse move on screen, does the capslock LED respond on the keyboard, are there any unusual LED indications from the front panel; e.g. disk LED jammed on.

There are some really interesting methods to capture data about a hang, but not sure if you want to go there yet. My favourite is a serial console with a laptop next to the failing system.

Seems to be a well known bug in Ubuntu Jaunty! See these threads http://ubuntuforums.org/showthread.php?t=1135055 https://bugs.launchpad.net/ubuntu/%2Bsource/linux/%2Bbug/355155 https://bugs.launchpad.net/ubuntu/%2Bsource/linux/%2Bbug/375986 a lot to read. I solved my problem be recompiling kernel 2.6.27 (see here for a guide http://ubuntuforums.org/showthread.php?t=311158 ) but had yet no luck to do the same with the real time kernel from Jaunty.

If the mouse still moves then you may be able to try giving a Control/Alt/Backspace to see if you can restart the X Server and login again, that does depend on how dead the system is but mouse motion means that you may also have keyboard control to some extent. Regards, Nick. ________________________________ Windows Live™: Keep your life in sync. Check it out! _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user

Unfortunately, I didn't ping the machine because it suddenly occurred to me that I'd never tried to ping the machine when it was working, so any result might be meaningless.

I just yanked the sound card and went back to on-board sound. I have a feeling that the system is going to be stable from here on out. It happened just that way once before. The only question, then, is is it a hardware or driver issue? I really don't care anymore.

Darren Landrum wrote: I take it all back. I was playing Leo Laporte's BitGravity stream when X suddenly crashed and restarted itself, and the mouse pointer was gone when I logged back in. So I had to restart the computer. Still, at least that's a new behavior. Also, Youtube videos are now acting weird. Either the video works, or the audio works, but not both at the same time. I'm about ready to take this entire machine out back and have at it with my sledgehammer.

Don't buy it if you want to use Linux. MOTU is Linux hostile and won't cooperate towards a driver. The best you will get is a reverse engineered driver, but currently nobody is working on the mk3. In any case, as it is now MOTU is evil.

On Sat, May 23, 2009 at 09:50:45PM -0400, Darren Landrum wrote: Ah, good, that reduces the component set to something quite smaller. The things I would consider as cause for that are: 1. dust in the CPU cooling fins, (I take my systems outside and run a vacuum cleaning in reverse, with a thin nozzle, and play the air stream over the various parts of the heatsink), 2. non-rotation or slow rotation of the cooling fans, (if the system has a BIOS sensor display, check that it shows a reasonable rotation of the fan, typical rotation rates are from 1000 to 3000 RPM, in my experience, and is fixable by replacing the fan, or cleaning it) 3. drying out of the thermal conducting grease between the CPU and the heatsink, (I recently had to remove and reapply the grease on a Pentium 4 3GHz desktop at home, symptom was CPU temperature consistently high and random power downs), 4. failing power supply, (I unplug non-essential devices temporarily, such as hard drive, to lower the average power draw, and see if the memtest symptom goes away ... I also check the power supply voltages with a meter), 5. corrosion or other damage to the memory DIMM socket or module, (I wiggle the DIMMs during a memtest, with about the equivalent of up to 200 gram force ... if the memtest result changes in a cycle with my wiggling, I know there's damage), 6. a specific memory DIMM failed, (remove it, see if memtest completes). Oh, and above all remember to use anti-static procedures, and try not to unplug or replug things inside the unit while the power is on. Static discharge damage is particularly annoying because it typically happens months after the static discharge happens. The discharge causes damage which then takes a long time before it begins to make the component fail. So "it works after I zapped it" isn't a reliable method of proving no damage was done.

Not an overhaul, but probably a good checkup and cleaning. I run machines older than that, but make sure to regularly open the cases up and clean up guck on fans and such. And to run it with the cover off and make sure that all the fans are working. I also take those little desiccant bags (that practically every piece of electronic gadgetry comes with these days) and pile them in the bottom of the case. I think that doing so has kept my systems running longer, keeps humidity down in the case. Darren Landrum wrote: > Overheating CPU? That's a thought. > > Well, I guess my system is a year old and could use a bit of an > overhaul. Thank you very much for the help! > > -- Darren > > > James Cameron wrote: >> On Sat, May 23, 2009 at 09:50:45PM -0400, Darren Landrum wrote: >>> Well, I decided to run memtest86, and it locked up during the test. >> Ah, good, that reduces the component set to something quite smaller. >> The things I would consider as cause for that are: >> >> 1. dust in the CPU cooling fins, (I take my systems outside and run a >> vacuum cleaning in reverse, with a thin nozzle, and play the air stream >> over the various parts of the heatsink), >> >> 2. non-rotation or slow rotation of the cooling fans, (if the system >> has a BIOS sensor display, check that it shows a reasonable rotation of >> the fan, typical rotation rates are from 1000 to 3000 RPM, in my >> experience, and is fixable by replacing the fan, or cleaning it) >> >> 3. drying out of the thermal conducting grease between the CPU and the >> heatsink, (I recently had to remove and reapply the grease on a Pentium >> 4 3GHz desktop at home, symptom was CPU temperature consistently high >> and random power downs), >> >> 4. failing power supply, (I unplug non-essential devices temporarily, >> such as hard drive, to lower the average power draw, and see if the >> memtest symptom goes away ... I also check the power supply voltages >> with a meter), >> >> 5. corrosion or other damage to the memory DIMM socket or module, (I >> wiggle the DIMMs during a memtest, with about the equivalent of up to >> 200 gram force ... if the memtest result changes in a cycle with my >> wiggling, I know there's damage), >> >> 6. a specific memory DIMM failed, (remove it, see if memtest >> completes). >> >> Oh, and above all remember to use anti-static procedures, and try not to >> unplug or replug things inside the unit while the power is on. >> >> Static discharge damage is particularly annoying because it typically >> happens months after the static discharge happens. The discharge causes >> damage which then takes a long time before it begins to make the >> component fail. >> >> So "it works after I zapped it" isn't a reliable method of proving no >> damage was done.

Indeed. Replace or at least re-seat the RAM.

On Sat, May 23, 2009 at 08:11:45PM -1000, david wrote: Fascinating. How have you measured the result?

Desiccant really only works properly when it is in a sealed enclosure, at other times it simply takes up the atmospheric water until it is full. I guess what you're really achieving is momentary bursts of lower humidity. Typical systems are specified with a minimum operating relative humidity of 10%, so removing all humidity may mean you're operating outside the specifications. Which means either the manufacturer doesn't see the point of testing "what happens if a pile of desiccant is placed in the case", or *has* tested it and decided it is bad. I can think of one advantage of your practice ... if the system is powered off each day, then the thermal cycle experienced by the desiccant may cause it to grab available moisture when the system is initially powered off, and release it slowly during the night into the room, and then release a final burst when the system is initially powered on. Now I'm imagining automatic opening and closing vents ... ;-) ... which if they were under software control could be tied into a performance.

david wrote: You can get cases with top mounted fans. The Antec 1200 has a 200mm fan on top, and 5 other fans!

Or for a real open airflow, maybe the Antec Skeleton. http://www.antec.com/Believe_it/product.php?id=NzA0 :)

Pieter Palmers wrote: Well, given my experience so far under Linux, maybe switching back to Windows isn't such a bad idea after all. Maybe I'd finally get some music made. Of course, if these are hardware issues, I can't really blame Linux for that. I can blame Linux for its lack of direction and major fragmentation, but that's just me being a cynic. There has to be something that works well under Linux that isn't M-Audio mediocrity. If the only choices are M-Audio (cheap, and sounds like it) and RME (way too expensive), then I think I'd rather spend my money on good software that will help me make music. -- Darren _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user

On Sat, May 23, 2009 at 12:12:02PM -0400, Darren Landrum wrote: It wouldn't happen to be an HDA-Intel sound card, would it? -ken _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user

Then it'll happen twice in a day. Can't imagine that it's good for the hard drive with all these hard restarts.

On Sun, 24 May 2009 21:57:18 -0700 "Aaron L." <elmastero74(a)gmail.com> wrote: Sounds similar to what happens to me (except the flash/firefox thing). It happens with graphics heavy stuff and during heavy compiles (kernel..). Mouse is only sometimes responsive for a few seconds. thought it was gone with 2.6.29 but it came back after a few weeks.. I'm pretty sure it's hardware related, overheating or something with graphics since it also happened when I had installed windows a few years ago to play at a LAN-party. Now it starts to get annoying, I'm just waiting for the perfect laptop that can be used as audio workstation.. --Philipp _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user

Ken, I have an Intel soundcard: aaron@mycomputername:~$ lspci |grep -e Audio 00:1f.5 Multimedia audio controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) AC'97 Audio Controller (rev 02) I'm hosed huh.

On Sun, May 24, 2009 at 6:41 PM, Ken Restivo <ken(a)restivo.org> wrote: > On Sat, May 23, 2009 at 12:12:02PM -0400, Darren Landrum wrote: > > Just had another lockup, a hard one (not even the mouse pointer would > > move) In addition (and this always happens, I forgot to mention it), the > > audio in the sound buffer at the time of the lock-up will loop over and > > over again making a sort of jack-hammer noise depending on what was in > > there at the time. > > > > This time, I had no proprietary Nvidia driver going, so that's > > eliminated as the cause. Unfortunately, I didn't ping the machine > > because it suddenly occurred to me that I'd never tried to ping the > > machine when it was working, so any result might be meaningless. I'll be > > sure to capture that baseline today. > > > > The other things I've eliminated is Flash playback, as this happened > > when I was playing back a long mp3 (a BBC podcast) in Totem. One other > > pattern I've noticed is that these hangs always happen when I'm playing > > back something with audio. I guess the next logical test is to go a week > > or so without playing anything and see what happens. > > > > So, I think it has to be the sound card. Whether it's a driver issue or > > a hardware issue remains to be seen. > > > > You can consider this email an angry rant more than anything, I suppose. > > :-/ Thanks for the help. > > > > It wouldn't happen to be an HDA-Intel sound card, would it? > > -ken > _______________________________________________ > Linux-audio-user mailing list > Linux-audio-user(a)lists.linuxaudio.org > http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user >

_______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user