Trouble brewing

[kilaueabart]

Lately I've noticed when shutting down Ubuntu I will get a few lines, or several screenfuls of lines ending with something like "(Transmitter ID)." And now I have noticed that my syslog is almost entirely filled with

Code: Select all

Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER: Corrected error received: 1000:00:00.0
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Transmitter ID)
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER:   device [1957:0421] error status/mask=00001000/00002000
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER:    [12] Timeout
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER: Corrected error received: 1000:00:00.0
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Transmitter ID)
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER:   device [1957:0421] error status/mask=00001000/00002000
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER:    [12] Timeout

Many of these lines are written every second. syslog when I checked it this morning was 63783 lines (3.3 MB) long. The above lines are now in syslog.1, 796086 lines, 86.0 MB, long, all from yesterday and today.
Until I checked back today, I was worried that my hard drive was going to be crammed full of syslogs, but apparently it knows when to quit adding to them.
I still wonder if the computing time lost to writing the log might not be contributing to the frequent slowing down and stalls I get on the computer.
Can anyone give me a clue how to fix this?

[kilaueabart]

Another problem I have occasionally is getting an odd-looking, dead Ubuntu Desktop. If the screenshot enlarges properly, notice the top and bottom panels, different images and different colors from the normal.

baddesktop.jpg

I rebooted right away when that happened this morning and then I checked auth.log (no idea why I picked that particular file) and was able to compare notations from the two boots. The failed one was shorter, but ended with the following lines. The successful boot lack those lines but went on quite a bit longer.

Mar 17 09:03:59 X5000 dbus[4219]: [system] Failed to activate service 'org.freedesktop.RealtimeKit1': timed out
(Above line repeated three times, followed by a line common to both boots)
Mar 17 09:05:14 X5000 dbus[4219]: [system] Failed to activate service 'org.freedesktop.RealtimeKit1': timed out
(Above line repeated six more times)

Does anyone see anything significant there ?

[xeno74]

Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER: device [1957:0421] error status/mask=00001000/00002000
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER: [12] Timeout
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER: Corrected error received: 1000:00:00.0
Mar 16 09:34:06 X5000 kernel: pcieport 1000:00:00.0: AER: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Transmitter ID)

Maybe the problem is a C-Media PCI audio card. Please reduce the RAM with the boot argument "mem=3500M" and check the log again.

Thanks

[kilaueabart]

Maybe the problem is a C-Media PCI audio card. Please reduce the RAM with the boot argument "mem=3500M" and check the log again.

Thank you, but I think I already have that, unless in the wrong place. It's right after "rootdelay=5" in the first section followed by a semi-colon. I fear my problem is something else.

Today I noticed a line that occurs now and then that wasn't included in the batch I sent:

Mar 18 14:24:05 X5000 kernel: pcieport 1000:00:00.0: AER: can't find device of ID0000

[xeno74]

Which kernel do you use?

[kilaueabart]

Which kernel do you use?

A 9 Dec version of 5.5. Haven't tested 5.6 yet.

[kilaueabart]

Others with a similar problem have been advised to turn off ASPM:
"we were able to solve the problem by setting the kernel parameter pcie_aspm=off."
Doing so meant messing around in GRUB which of course I ain't got.

[xeno74]

Others with a similar problem have been advised to turn off ASPM:
"we were able to solve the problem by setting the kernel parameter pcie_aspm=off."
Doing so meant messing around in GRUB which of course I ain't got.

You can add “pcie_aspm=off“ to the boot arguments in U-Boot.

Code: Select all

setenv bootargs root=/dev/sdd2 mem=3500M pcie_aspm=off

[kilaueabart]

You can add “pcie_aspm=off“ to the boot arguments in U-Boot.

Code: Select all

setenv bootargs root=/dev/sdd2 mem=3500M pcie_aspm=off

Thank you. That saved the time and disc space all those messages were consuming.

But I wonder whether that just means errors that I started getting a month or two ago are just not being reported any more. Is there something that really needs fixing?