Laggy & Finally Completely Frozen 12.04.2 on Large cp or rsync
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
linux (Ubuntu) |
Expired
|
High
|
Unassigned |
Bug Description
I have two pairs of raided disks. I am trying to copy 1/2 T of data from one to the other. First I tried "cp /from /to". It seemed to behave ok, so I left it overnight. When I came back in the AM, the machine was completely frozen. I had to use the power switch. Only about 150GB had been copied. I tried again with "nice -n 10 cp ..." and the same thing happened. Then I discovered ionice and shifted to rsync. So I used "ionize -n 7 rsync -cva /from /to". I also turned off the sleep/lock for the display. Again it seemed to behave ok, so I left it while I went for dinner. Two-three hours later, I found that all windows were frozen. I couldn't use REISUB, the reset switch _or_ the power switch. I had to power off using the switch on the power supply. Rebooting, I found that about 300G had been copied by this point. I am trying for the 4th time now, again using rsync. iotop is running. It regularly freezes between rsync's copying one file and the next. If I try to shift to a window or type at the terminal, there is a freeze until the file being copied by rsync is finished. If I keep typing at the terminal (or in the text-box in Firefox, as I am now) then things become smooth until I stop for a bit. Then the freezing begins to happen again. (That's not completely true. Tab completion at the terminal is a disaster. Cut & paste in this window is the same.) It may be my fantasy, but it seems that if I keep using the machine, getting it to respond to keyboard/mouse intput, then it doesn't permanently freeze up. It's when I leave to it's own devices for some time that I have a problem.
The only difference between the last try and the first 3, is that after having a look at dmesg, I enabled IOMMU in the BIOS. Probably irrelevant, but I did also notice the "spurious 8259A interrupt: IRQ7" message in dmesg that has been reported in bug #75626.
I haven't changed anyting with the I/O scheduler, which has been reported as causing I/O problems in bug# 427210 and #131094. So, it still reads: "noop [deadline] cfq"
Tyan S3970 with 2 CPU's (8 cores).
Two pairs of WD drives (500G & 2T)
TARO add-on storage card (used by the 2T drives)
Ubuntu 12.04.3 fully updated.
Essentially new install (During this process I only added iotop, atop, htop, acpi, m5deep & emacs)
500G drives are RAID-1; 2T drives are RAID-1 & encrypted.
ProblemType: Bug
DistroRelease: Ubuntu 12.04
Package: linux-image-
ProcVersionSign
Uname: Linux 3.5.0-39-generic x86_64
NonfreeKernelMo
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.25.
ApportVersion: 2.0.1-0ubuntu17.4
Architecture: amd64
AudioDevicesInUse:
USER PID ACCESS COMMAND
/dev/snd/
/dev/snd/
CRDA: Error: [Errno 2] No such file or directory
Card0.Amixer.info:
Card hw:0 'NVidia'/'HDA NVidia at 0xfe77c000 irq 17'
Mixer name : 'Nvidia GPU 0b HDMI/DP'
Components : 'HDA:10de000b,
Controls : 24
Simple ctrls : 4
Card1.Amixer.info:
Card hw:1 'Audigy2'/'SB Audigy 2 ZS [SB0353] (rev.4, serial:0x10031102) at 0xd880, irq 29'
Mixer name : 'SigmaTel STAC9721,23'
Components : 'AC97a:83847609'
Controls : 210
Simple ctrls : 46
Date: Tue Aug 27 19:30:21 2013
HibernationDevice: RESUME=
InstallationMedia: Ubuntu 12.04.2 LTS "Precise Pangolin" - Release amd64 (20130214)
IwConfig:
eth0 no wireless extensions.
eth1 no wireless extensions.
lo no wireless extensions.
MachineType: empty empty
MarkForUpload: True
ProcEnviron:
TERM=xterm
PATH=(custom, no user)
LANG=en_US.UTF-8
SHELL=/bin/bash
ProcFB:
ProcKernelCmdLine: BOOT_IMAGE=
RelatedPackageV
linux-
linux-
linux-firmware 1.79.6
RfKill:
SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 11/20/2008
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 'V2.05 '
dmi.board.
dmi.board.name: S3970
dmi.board.vendor: TYAN Computer Corporation
dmi.board.version: empty
dmi.chassis.
dmi.chassis.type: 3
dmi.chassis.vendor: empty
dmi.chassis.
dmi.modalias: dmi:bvnAmerican
dmi.product.name: empty
dmi.product.
dmi.sys.vendor: empty
---
ApportVersion: 2.12.1-0ubuntu2
Architecture: amd64
AudioDevicesInUse:
USER PID ACCESS COMMAND
/dev/snd/
/dev/snd/
CasperVersion: 1.336
DistroRelease: Ubuntu 13.10
IwConfig:
eth0 no wireless extensions.
eth1 no wireless extensions.
lo no wireless extensions.
LiveMediaBuild: Ubuntu 13.10 "Saucy Salamander" - Alpha amd64 (20130827)
MachineType: empty empty
MarkForUpload: True
Package: linux (not installed)
ProcFB: 0 nouveaufb
ProcKernelCmdLine: file=/cdrom/
ProcVersionSign
RelatedPackageV
linux-
linux-
linux-firmware 1.113
RfKill:
Tags: saucy
Uname: Linux 3.11.0-4-generic x86_64
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo
dmi.bios.date: 11/20/2008
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 'V2.05 '
dmi.board.
dmi.board.name: S3970
dmi.board.vendor: TYAN Computer Corporation
dmi.board.version: empty
dmi.chassis.
dmi.chassis.type: 3
dmi.chassis.vendor: empty
dmi.chassis.
dmi.modalias: dmi:bvnAmerican
dmi.product.name: empty
dmi.product.
dmi.sys.vendor: empty
tags: |
added: needs-crash-log regression-potential removed: freezing heavy io |
Changed in linux (Ubuntu): | |
status: | Confirmed → Incomplete |
Changed in linux (Ubuntu): | |
importance: | Undecided → High |
tags: | added: latest-bios-unknown |
tags: | added: needs-upstream-testing |
Ok, on this 4th go-round, while using firefox to look at this bug report, it killed all processes & logged me out. I was able to log back in, and found that I have only succeeded in copying 375G.