Disco currently has the e1000e fix in disco-proposed (version 5.0.0-30.32),
also available as linux-hwe kernel in bionic-proposed.
** Changed in: linux (Ubuntu Disco)
Status: In Progress => Fix Committed
--
You received this bug notification because you are a member of Kernel
Packages, wh
Hi Drew,
Yes, the HWE kernel syncs automatically from the normal kernel as it
moves forward.
Once Disco gets the patch, the HWE from Disco in Bionic should get it as
well (same version number plus ~18.04.1 suffix).
--
You received this bug notification because you are a member of Kernel
Package
Marking Bionic as Fix Released as the kernel from bionic-proposed has
been promoted to bionic-updates.
** Changed in: linux (Ubuntu Bionic)
Status: Fix Committed => Confirmed
** Changed in: linux (Ubuntu Bionic)
Status: Confirmed => Fix Released
--
You received this bug notificati
Marking Bionic as Fix Released as the kernel from bionic-proposed has
been promoted to bionic-updates.
** Changed in: linux (Ubuntu Bionic)
Status: Fix Committed => Fix Released
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux
Marking status on B/X/D as Incomplete.
(email below sent to kernel-team mailing list
as replies to both patch series above).
Please hold / don't apply this patch for now.
The reporter hit an apparently unrelated Oops in 3 of 40 nodes,
and it hasn't been possible yet to determine whether this pat
Attaching the debdiffs for zfs-linux/spl-linux on X/B/D/E,
for documentation purposes; will send testing/notes later.
Independently of the kernel packaging approach determined
to enable debug symbols on ZFS/SPL modules, these kind of
patches for the userspace packages are be required anyway,
and c
Hi Martijn,
Thanks for testing bionic-proposed!
So it will be resolved for bionic kernels shortly, when it hit bionic-updates.
Disco/19.04 will get this patch via stable updates in the near future
[1].
Eoan has it applied (LP: #1837725).
So this is all good.
Thanks again,
Mauricio
** Also aff
[Unstable][PATCH 0/6] Add support for ZFS debug symbols
https://lists.ubuntu.com/archives/kernel-team/2019-August/103425.html
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1840704
Title:
Test Build 3) New behavior if ZFS modules are built with debug symbols
goal: show zfs debug symbols are correctly built and packaged into non-
debug & debug packages.
- test packaging
- zfs built with debug symbols
- zfs modules present in debug package
- extra modules *have* .gnu_debuglink secti
Test Build 2) New behavior if ZFS modules are *not* built with debug
symbols
goal: show failsafe/backwards compatible behavior if zfs-dksm doesn't
support/build debug symbols
and kernel build log reports missing debug symbols, and extra modules
have .gnu_debuglink.
- test packaging
- zfs
Test Build 4) All debug symbols disabled
goal: show no zfs debug symbol activity happens either (along w/ other
debug symbol stuff)
- test packaging
- nothing built with debug symbols
- no debug package present
- no .gnu_debuglink section at all
- (no regressions)
Test packaging, debug symbols
Test Build 1) Old behavior
goal: show limitations/issues.
- original packaging
- zfs not built with debug symbols
- zfs modules not present in debug package
- extra modules lack .gnu_debuglink section
Original packaging:
There are no ZFS modules in the debug package:
$ dpkg-deb -x linu
** Description changed:
The ZFS kernel modules aren't built with debug symbols,
which introduces problems/issues for debugging/support.
- Patches will be sent soon for linux and zfs/spl-linux,
- covering X/B/D/E/Unstable.
+ Patches are required in:
+ 1) linux kernel packaging, to add infras
** Changed in: linux (Ubuntu Eoan)
Status: Incomplete => Fix Released
** Changed in: linux (Ubuntu Disco)
Status: New => In Progress
** Changed in: linux (Ubuntu Bionic)
Status: New => Fix Committed
--
You received this bug notification because you are a member of Kernel
Pa
Drew,
Thanks for testing bionic-proposed!
So it will be resolved for bionic kernels shortly, when it hit bionic-updates.
Disco/19.04 will get this patch via stable updates in the near future
[1].
Eoan has it applied (LP: #1839588).
So this is all good.
Thanks again,
Mauricio
[1] https://lists
Hi Drew,
That's very good news! So it looks like that patch resolves the
problem.
Could you please test the kernel in bionic-proposed [1] (4.15.0-60-generic)
which has that patch to confirm it's also working correctly?
Thanks!
Mauricio
[1] https://wiki.ubuntu.com/Testing/EnableProposed
--
Yo
Hi Martin,
There's a potential fix for this upstream, in v5.3-rc1 mainline build (thus not
in v5.2),
which is also applied to bionic-proposed (4.15.0-60.67).
Could you please test whether the issue is resolved with bionic-proposed
[1] ?
If that doesn't help, further regression/bisect test steps
Mentioned upstream candidate fix is:
commit df9a606184bfdb5ae3ca9d226184e9489f5c24f7
Author: Suganath Prabu
Date: Tue Jul 30 03:43:57 2019 -0400
scsi: mpt3sas: Use 63-bit DMA addressing on SAS35 HBA
Although SAS3 & SAS3.5 IT HBA controllers support 64-bit DMA addressing, as
pe
Hi Drew,
There's a mpt3sas fix in v5.3-rc3 for a problem that may cause an adapter
firmware fault
(although not sure of the exact fault state code; but it should cause a reset
anyway).
If you could please test either
1) v5.3-rc2 [1] to confirm the issue happens with v5.3-rc2 but not with
v5.3-
[X/B][PATCH] bnx2x: Disable multi-cos feature.
https://lists.ubuntu.com/archives/kernel-team/2019-August/103282.html
[D][PATCH] bnx2x: Disable multi-cos feature.
https://lists.ubuntu.com/archives/kernel-team/2019-August/103283.html
--
You received this bug notification because you are a member o
Somewhat similarly on recent 5.2 kernel without the fix.
(again with options to ignore hardware errors/faults)
Aug 19 17:15:15 HOSTNAME kernel: Uhhuh. NMI received for unknown reason 21 on
CPU 0.
Aug 19 17:15:15 HOSTNAME kernel: perf interrupt took too long (3222 > 2500),
lowering kernel.perf_ev
Older crashdump analysis confirmed the bnx2x driver/status
being in traffic class setup / stop hardware in LLDP path.
PID: 3936 TASK: 883fdc9b1c00 CPU: 11 COMMAND: "kworker/11:0"
#0 [883fec593ce0] __schedule at 81850bae
#1 [883fec593d30] schedule at 818510f5
#2 [
For documentation purposes, in a recent Xenial/4.4 kernel,
this kernel error log is seen (with options to ignore the
hardware error/fault that panics/reboots the system).
[ 113.658876] bnx2x: [bnx2x_stats_comp:205(eno1)]timeout waiting for stats
finished
[ 123.648066] bnx2x: [bnx2x_state_wait:3
ay into stable kernels
+(e.g., v5.2.8 which made Eoan/19.10 recently).
** Also affects: linux (Ubuntu Bionic)
Importance: Undecided
Status: New
** Also affects: linux (Ubuntu Xenial)
Importance: Undecided
Status: New
** Also affects: linux (Ubuntu Eoan)
Importance: Unde
Verification successful with bionic-proposed.
No warning nor oops anymore.
# uname -rv
4.15.0-59-generic #66-Ubuntu SMP Wed Aug 14 10:56:44 UTC 2019
# ./setup-bcache-wb_percent-before-attach.sh >/dev/null 2>&1
[ 105.696881] bcache: register_bdev() registered backing device loop0
[ 105.703809] b
Public bug reported:
Description/patches to be provided this week.
** Affects: linux (Ubuntu)
Importance: Undecided
Assignee: Mauricio Faria de Oliveira (mfo)
Status: In Progress
** Changed in: linux (Ubuntu)
Status: New => In Progress
** Changed in: linux (Ubu
ricio Faria de Oliveira (mfo)
Status: In Progress
** Changed in: linux (Ubuntu)
Status: New => In Progress
** Changed in: linux (Ubuntu)
Assignee: (unassigned) => Mauricio Faria de Oliveira (mfo)
--
You received this bug notification because you are a member of Kern
Verification done for linux on xenial-proposed.
The inodes for file, xattr dir, and xattr child are all evicted at file
removal time, not making it to any disposal list after file removal.
So the window/scenario for the problem to occur is not present anymore.
Log
---
$ uname -rv
4.4.0-160-gene
Verification successful with disco-proposed.
No warning nor oops anymore.
# rmadison -s disco-proposed linux
linux | 5.0.0-26.27 | disco-proposed | source
# uname -rv
5.0.0-26-generic #27-Ubuntu SMP Tue Aug 13 17:47:39 UTC 2019
# ./setup-bcache-wb_percent-before-attach.sh >/dev/null 2>&1
[ 171
Hi @mathew-hodson,
This fix is needed in the Linux kernel package for Xenial as well (it
duplicates the zfs-linux/dkms source).
Shouldn't the 'linux' source package still be tracked?
Thanks,
Mauricio
--
You received this bug notification because you are a member of Kernel
Packages, which is su
Verification done for zfs-linux on xenial-proposed.
The inodes for file, xattr dir, and xattr child are all evicted at file
removal time, not making it to any disposal list after file removal.
So the window/scenario for the problem to occur is not present anymore.
Log
---
$ echo 'deb http://arc
Colin, thanks for sponsoring and additional testing!
Is this ZFS regression test suite available to other people?
It sounds great for future work with ZFS.
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.laun
id
** Changed in: linux (Ubuntu Xenial)
Status: New => In Progress
** Changed in: linux (Ubuntu Xenial)
Assignee: (unassigned) => Mauricio Faria de Oliveira (mfo)
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to zfs-lin
Fix Change Details
--
(Compare with comment #3)
The xattr dir/child inodes are evicted/disposed
immediately on file removal, not on shrinker.
$ touch /zfs/file
<...>-23940 [006] d... 10808.746798: p_zfs_mknode_0:
(zfs_mknode+0x0/0xe00 [zfs]) flag=0x0 dzp=0x88011a
zfs-test results w/ original & modified zfs-dkms
original
$ uname -rv
4.4.0-159-generic #187-Ubuntu SMP Thu Aug 1 16:28:06 UTC 2019
$ modinfo zfs | head
filename: /lib/modules/4.4.0-159-generic/updates/dkms/zfs.ko
version:0.
zfs-test test suite on xenial
-
Build/install the latest zfs-test suite from zfs.git:
$ git clone https://github.com/zfsonlinux/zfs.git
$ cd zfs
$ git log --oneline -1
c81f179 Metaslab max_size should be persisted while unloaded
$ sudo apt-get build-dep zfs-linux
$ su
has been run on original/modified
zfs-dkms package/kernel modules, with no regressions.
** Affects: zfs-linux (Ubuntu)
Importance: Undecided
Status: Invalid
** Affects: zfs-linux (Ubuntu Xenial)
Importance: Undecided
Assignee: Mauricio Faria de Oliveira (mfo)
Stat
** Attachment added: "zfs-kprobes.sh"
https://bugs.launchpad.net/ubuntu/+source/zfs-linux/+bug/1839521/+attachment/5281633/+files/zfs-kprobes.sh
** Attachment removed: "kmod-zget.c"
https://bugs.launchpad.net/ubuntu/+source/zfs-linux/+bug/1839521/+attachment/5281631/+files/kmod-zget.c
--
Test-case for the problem (kmod-zget.c).
Header comments with steps to reproduce.
** Attachment added: "kmod-zget.c"
https://bugs.launchpad.net/ubuntu/+source/zfs-linux/+bug/1839521/+attachment/5281631/+files/kmod-zget.c
--
You received this bug notification because you are a member of Kerne
debdiff for zfs-linux (zfs-linux_shrinker-deadlock-xattr.debdiff)
based on ubuntu/xenial-devel branch from LP source pkg git repo [1].
[1] https://git.launchpad.net/ubuntu/+source/zfs-linux
** Patch added: "zfs-linux_shrinker-deadlock-xattr.debdiff"
https://bugs.launchpad.net/ubuntu/+source/z
Problem Summary:
---
If an xattr directory inode and its xattr child inode are on the _same_
disposal list,
and the xattr directory inode is _before_ its xattr child inode in this
disposal list...
Then zfs_purgedir() of the xattr directory calls zfs_zget() for the xattr child
inode
Problem Details:
---
Helper scripts
(create /zfs.img and mount it on /zfs; and setup kprobe events for debug)
$ sudo ./zfs-mount.sh
$ sudo ./zfs-kprobes.sh
Print kprobe events to screen as we go:
$ sudo cat /sys/kernel/debug/tracing/trace_pipe &
Create file:
- allocates normal/fil
** Attachment added: "kmod-zget.c"
https://bugs.launchpad.net/ubuntu/+source/zfs-linux/+bug/1839521/+attachment/5281634/+files/kmod-zget.c
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to zfs-linux in Ubuntu.
https://bugs.launchpad.net/
Fix Details
---
Problem report (GH Issue) upstream in ZFS on Linux:
https://github.com/zfsonlinux/zfs/issues/4816 ("Looping in zfs_zget()")
With very similar stack trace (syscall -> shrink ->
zfs_{inactive/rmnode/purgedir/zget})
https://github.com/zfsonlinux/zfs/issues/48
** Attachment added: "zfs-mount.sh"
https://bugs.launchpad.net/ubuntu/+source/zfs-linux/+bug/1839521/+attachment/5281632/+files/zfs-mount.sh
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to zfs-linux in Ubuntu.
https://bugs.launchpad.ne
[Original Description]
One LXC user reported lots of processes stuck in D state:
several threads waiting in the memory shrinker semaphore
(this symptom was thought to be/fixed via LP bug 1817628).
After some time, a provided crashdump revealed the issue
in ZFS's evict node path running in memory
Verification done on the new adapter, Solarflare X2542.
The tester reports that iperf3 tests ran solid for 24h-72h.
** Tags removed: verification-needed-bionic
** Tags added: verification-done-bionic
--
You received this bug notification because you are a member of Kernel
Packages, which is sub
The feature documentation in the Server Guide has been published [1].
[1] https://help.ubuntu.com/lts/serverguide/advanced-
installation.html#iscsi
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.ne
Verification done for bionic-proposed
based on the 3 tests / comments above.
** Tags removed: verification-needed-bionic
** Tags added: verification-done-bionic
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs
Verification with bionic-proposed of fio results.
For details see comment #5.
Overall performance has improved on the -proposed kernel
(tested in build without the patchset) and the patchset
as is in the -proposed kernel did improve it a bit more.
The results below show 3 results for each cache-
Verification with bionic-proposed of xfstests results.
No regressions introduced by this bcache patchset.
The comparison between -updates and -proposed is not directly possible
because -proposed introduced failures via other components in I/O path
(e.g., block, ext4).
This is described below, an
Verification done for Disco (one patch change only).
Only one of the two bcache devices stop working upon failures in one backing
device.
(see comment #21 for details).
# uname -rv
5.0.0-22-generic #23-Ubuntu SMP Tue Jul 23 17:23:54 UTC 2019
# ./setup-two-bcache-one-cache.sh >/dev/null 2>&1
[
Verification with bionic-proposed of the I/O Error path.
All good, working as expected (see comments #11 to #16).
# uname -rv
4.15.0-56-generic #62-Ubuntu SMP Wed Jul 24 20:18:55 UTC 2019
test 1
--
# ./setup.sh >/dev/null 2>&1
[ 369.375820] bcache: register_bdev() registered backing devic
Regression testing done on an older/previously supported adapter, SFC
7000 series.
The netperf suite of TCP/UDP STREAM and RR, and TCP_RR ran for ~2 days,
with results in the same ballpark as the original kernel and test kernels.
Now waiting for test results with the new/requested adapter
before
Patches sent to the kernel-team mailing list:
[B][PATCH] bcache: only set BCACHE_DEV_WB_RUNNING when cached device attached
https://lists.ubuntu.com/archives/kernel-team/2019-July/102653.html
[D][PATCH] bcache: only set BCACHE_DEV_WB_RUNNING when cached device attached
https://lists.ubuntu.com/ar
Tested with disco-proposed + patch (problem does not happen)
---
# uname -rv
5.0.0-22-generic #23+test20190725b1 SMP Mon Jul 29 14:43:55 -03 2019
# ./setup-bcache-wb_percent-before-attach.sh >/dev/null 2>&1
[ 69.567775] bcache: register_bdev() registered backing device loop0/null
2>&1
[ 69.
Attaching test-case script.
** Attachment added: "setup-bcache-wb_percent-before-attach.sh"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1837788/+attachment/5279850/+files/setup-bcache-wb_percent-before-attach.sh
--
You received this bug notification because you are a member of Kerne
Tested with bionic-proposed + patch (problem does not happen)
---
Test kernel available in:
https://people.canonical.com/~mfo/lp1837788/linux_4.15.0-56.62+test20190725b1_amd64/
# uname -rv
4.15.0-56-generic #62+test20190725b1 SMP Thu Jul 25 12:44:16 -03 2019
# ./setup-bcache-wb_percent-before
** Also affects: linux (Ubuntu Eoan)
Importance: Undecided
Assignee: Mauricio Faria de Oliveira (mfo)
Status: Confirmed
** Also affects: linux (Ubuntu Disco)
Importance: Undecided
Status: New
** Also affects: linux (Ubuntu Bionic)
Importance: Undecided
Status
Tested with eoan-proposed (problem does not exist)
root@bionic-bcache:~# uname -rv
5.2.0-9-generic #10-Ubuntu SMP Sat Jul 20 14:05:52 UTC 2019
# ./setup-bcache-wb_percent-before-attach.sh >/dev/null 2>&1
[ 54.032674] bcache: register_bdev() registered backing device loop0
[ 54.036972] bc
Tested with disco-proposed (problem exists)
-
# uname -rv
5.0.0-22-generic #23-Ubuntu SMP Tue Jul 23 17:23:54 UTC 2019
# ./setup-bcache-wb_percent-before-attach.sh >/dev/null 2>&1
[ 57.992319] bcache: register_bdev() registered backing device loop0/null 2>&1
[ 57.997331] bcache: run_cache
Tested with bionic-proposed (problem exists)
---
# uname -rv
4.15.0-56-generic #62-Ubuntu SMP Wed Jul 24 20:18:55 UTC 2019
# ./setup-bcache-wb_percent-before-attach.sh >/dev/null 2>&1
[ 39.096888] bcache: register_bdev() registered backing device loop0/null 2>&1
[ 39.109511] bcache: run_cache
For reference, this is the Warning in the CurrentDmesg.txt file.
[ 71.925684] WARNING: CPU: 8 PID: 2977 at
/build/linux-jmOa1Y/linux-4.15.0/drivers/md/bcache/writeback.c:603
bch_cached_dev_writeback_start+0x13f/0x1af [bcache]
[ 71.925686] Modules linked in: ebtable_filter ebtables ip6table_f
This is the warning trigger line:
int bch_cached_dev_writeback_start(struct cached_dev *dc)
...
WARN_ON(test_and_set_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags));
...
This commit fixes this problem:
commit 1f0ffa67349c56ea54c03ccfd1e073c990e7411e
Author: Coly Li
** Changed in: linux (Ubuntu)
Assignee: (unassigned) => Mauricio Faria de Oliveira (mfo)
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1837788
Title:
bcache kernel warning w
The documentation about the feature/parameters has been merged [1]
into Server Guide -> Installation -> Advanced Installaion -> iSCSI [2].
It's not yet published right now, so the HTML might be updated later.
cheers,
Mauricio
[1] https://code.launchpad.net/~mfo/serverguide/ibft/+merge/370264
[2]
Regression test results/log/script,
for documentation purposes.
** Attachment added: "lp1836635-test-regression.tar.xz"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1836635/+attachment/5277232/+files/lp1836635-test-regression.tar.xz
--
You received this bug notification because you a
[Bionic][PULL] sfc: patches for LP#1836635
https://lists.ubuntu.com/archives/kernel-team/2019-July/102196.html
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1836635
Title:
Bionic: suppo
** Description changed:
[Impact]
- * Support for Solarflare X2542 network adapter
-(Medford2 / SFC9250) in the sfc driver.
+ * Support for Solarflare X2542 network adapter
+ (Medford2 / SFC9250) in the sfc driver.
- * This network adapter is present on recent hardware,
-at le
Bionic)
Importance: Undecided
Assignee: Mauricio Faria de Oliveira (mfo)
Status: In Progress
** Affects: linux (Ubuntu Cosmic)
Importance: Undecided
Status: Invalid
** Affects: linux (Ubuntu Disco)
Importance: Undecided
Status: Invalid
** Affects
** Also affects: linux (Ubuntu Eoan)
Importance: Undecided
Assignee: Mauricio Faria de Oliveira (mfo)
Status: Invalid
** Also affects: linux (Ubuntu Disco)
Importance: Undecided
Status: New
** Changed in: linux (Ubuntu Disco)
Status: New => In Progress
** Chan
I/O Error Test 6 (for the Disco kernel)
commit: 'Revert "bcache: set CACHE_SET_IO_DISABLE in
bch_cached_dev_error()"'
Problem: if one backing device hits I/O errors the cache device
is disabled, but if that cache device is shared by other bcache
devices they stop too (even with n
[D/E][PATCH 0/1] LP#1829563 bcache: pending patch for Disco/Eoan
https://lists.ubuntu.com/archives/kernel-team/2019-July/102006.html
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1829563
[C][PATCH 0/2] LP#1829563 bcache: pending patches for Cosmic
https://lists.ubuntu.com/archives/kernel-team/2019-July/102001.html
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1829563
Titl
I/O Error Test 6 (for the Cosmic kernel)
commit: 'Revert "bcache: set CACHE_SET_IO_DISABLE in
bch_cached_dev_error()"'
Problem: if one backing device hits I/O errors the cache device
is disabled, but if that cache device is shared by other bcache
devices they stop too (even with
I/O Error Test 7
commit: "bcache: fix ioctl in flash device"
This test needs QEMU+GDB since there's no way to set the variable
from the kernel, because it references incorrect memory/pointer.
In the original kernel, the ioctl() to a flash-only volume
goes from ioctl_dev() down t
[B][PATCH 00/11] LP#1829563 bcache: risk of data loss on I/O errors in backing
or caching devices
https://lists.ubuntu.com/archives/kernel-team/2019-July/101989.html
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https:/
** Description changed:
[Impact]
- * The bcache code in Bionic lacks several fixes to handle
-I/O errors in both backing devices and caching devices.
+ * The bcache code in Bionic lacks several fixes to handle
+ I/O errors in both backing devices and caching devices.
- * Partial
I/O Error Test 1
commit "bcache: add CACHE_SET_IO_DISABLE to struct cache_set flags"
Problem: the cacheset is not retired immediately on I/O errors in
cache device if I/O requests keep coming.
Original kernel: bcache device remains on top of caching device,
and 'fio' never finis
I/O Error Test 4
commit "bcache: add io_disable to struct cached_dev"
Problem: in case of the backing device hits I/o errors or is
disconected, the I/O can still be accepted to the bcache device.
Original kernel: dd writes in writeback mode to failed backing device complete.
Mod
I/O Error Test 2
commit "bcache: add stop_when_cache_set_failed option to backing device"
Problem: upon too many io errors on cache device, bcache code retires cache set,
but it doesn't stop the bcache device.
(in writeback mode, if it continue to write, writes go to backing dev
I/O Error Test 5
commit "bcache: stop bcache device when backing device is offline"
Problem: bcache is unaware that a backing device goes offline,
and accepts writes.
Original kernel: bcache doesn't realize backing device is offline.
Modified kernel: bcache device is removed aft
I/O Error Test 6
Make sure that if two bcache devices share a cache device,
only one bcache device is offlined in case of I/O errors
in only one backing device.
Original
# uname -rv
4.15.0-55-generic #60-Ubuntu SMP Tue Jul 2 18:22:20 UTC 2019
# ./setup-two-bcache-one-
I/O Error Test 3
commit "bcache: fix inaccurate io state for detached bcache devices"
Probçem: on detached bcache devices, iostat reports non-zero 'avgqu-sz'
and '%util' even after IO stopped.
Original kernel: iostat fields are non-zero on I/O idle.
Modified kernel: iostat fiel
** Attachment added: "setup-two-bcache-one-cache.sh"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1829563/+attachment/5275535/+files/setup-two-bcache-one-cache.sh
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
** Attachment added: "setup.sh"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1829563/+attachment/5275533/+files/setup.sh
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1829563
** Attachment added: "setup-sda.sh"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1829563/+attachment/5275534/+files/setup-sda.sh
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1
Tests with I/O errors
=
The I/O error tests in the caching and backing device have been
simulated with the device mapper dm-linear and dm-error targets.
The tests are based on the patch messages, which describe a
problem and/or expected behavior, so to exercise the change.
Co
** Attachment added: "test-bcache-cache-modes.fio.sh"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1829563/+attachment/5275532/+files/test-bcache-cache-modes.fio.sh
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubunt
Tests with fio
==
The fio tool has been run for performance tests on the original
and modified kernels in sequential (rw) and random read/write (randrw)
on all cache modes available (writethrough, writeback, writearound, none).
Summary: no regressions observed, same IOPS and BW metri
** Attachment added: "test-bcache-cache-modes.xfstests.sh"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1829563/+attachment/5275531/+files/test-bcache-cache-modes.xfstests.sh
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linu
Tests with xfstests
===
The xfstests test suite has been run on the original
and modified kernels -- on all cache modes available
(writethrough, writeback, writearound, none).
Summary: no regressions observed; apparently one FIX
achieved in the writearound cache mode (generic/371)
Fred,
Glad to know it.
Thanks for the additional testing and feedback.
cheers,
Mauricio
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1817321
Title:
installer does not support iSCSI i
Hi Fred,
At this time it's only available in the netboot images (network
installs) [1].
It should be available on ISOs in Eoan release 19.10 and next
Bionic point-release 18.04.3 (currently scheduled to August [2]).
We'll put some documentation about the installer kernel/preseed parameters soon
** Description changed:
+ [Impact]
+
+ * The bcache code in Bionic lacks several fixes to handle
+I/O errors in both backing devices and caching devices.
+
+ * Partial or permanent errors in backing or caching devices,
+specially in writeback mode, can lead to data loss and/or
+the
** Also affects: linux (Ubuntu Cosmic)
Importance: Undecided
Status: New
** Also affects: linux (Ubuntu Bionic)
Importance: Undecided
Status: New
** Changed in: linux (Ubuntu Bionic)
Assignee: (unassigned) => Mauricio Faria de Oliveira (mfo)
** Changed in: linux (Ubu
Verified debian-installer in {disco,cosmic,bionic}-proposed.
The verification has been done with the netboot image files
for disco/cosmic/bionic (and hwe-netboot for bionic),
using regular and lvm partitioning,
on VMs for the architectures amd64/i386/arm64/ppc64el
and baremetal for amd64.
All tes
verification done for bionic/partman-iscsi.
Use of partman-iscsi/iscsi_auto correctly writes /etc/iscsi/iscsi.initramfs
with either ISCSI_AUTO=true or iSCSI LUN details with the right MAC address.
~ # grep 'retrieving partman-iscsi' /var/log/syslog
Jun 6 15:01:20 anna[1605]: DEB
verification done for cosmic/partman-iscsi.
Use of partman-iscsi/iscsi_auto correctly writes /etc/iscsi/iscsi.initramfs
with either ISCSI_AUTO=true or iSCSI LUN details with the right MAC address.
~ # grep 'retrieving partman-iscsi' /var/log/syslog
Jun 6 14:20:51 anna[1521]: DEBU
verification done for disco/partman-iscsi.
Use of partman-iscsi/iscsi_auto correctly writes /etc/iscsi/iscsi.initramfs
with either ISCSI_AUTO=true or iSCSI LUN details with the right MAC address.
~ # grep 'retrieving partman-iscsi' /var/log/syslog
Jun 6 14:04:03 anna[1582]: DEBUG
401 - 500 of 718 matches
Mail list logo
