f2fs/008: avoid endless wait
PATCH 1/1 f2fs/008: avoid endless wait
"udevadm wait <dev>" without a --timeout=SECONDS parameter will wait endlessly. Endless wait can be triggered in f2fs/008 by e.g. using a zram device as a SCRATCH_DEV, where "device type is unknown" failure sees the /dev/mapper/$vgname-$lvname node never appear.
---
It might make more sense to add a default timeout to _udev_wait(), but that could also call udev{adm }settle.
tests/f2fs/008 | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
tests/f2fs/008 +6 -3
@@ -35,9 +35,12 @@ _cleanup() rm -f $tmp.* } -$LVM_PROG pvcreate -f $SCRATCH_DEV >>$seqres.full 2>&1 -$LVM_PROG vgcreate -f $vgname $SCRATCH_DEV >>$seqres.full 2>&1 -$LVM_PROG lvcreate -y -L 1024m -n $lvname $vgname >>$seqres.full 2>&1 +$LVM_PROG pvcreate -f $SCRATCH_DEV >>$seqres.full 2>&1 \ + || _notrun "unabled to create LVM physical volume" +$LVM_PROG vgcreate -f $vgname $SCRATCH_DEV >>$seqres.full 2>&1 \ + || _notrun "unabled to create LVM volume group" +$LVM_PROG lvcreate -y -L 1024m -n $lvname $vgname >>$seqres.full 2>&1 \ + || _notrun "unabled to create LVM logical volume" _udev_wait /dev/mapper/$vgname-$lvname _mkfs_dev /dev/mapper/$vgname-$lvname >>$seqres.full 2>&1
Signature · 1 line
2.43.0
Quoted 9 lines · David Disseldorp
"udevadm wait <dev>" without a --timeout=SECONDS parameter will wait endlessly. Endless wait can be triggered in f2fs/008 by e.g. using a zram device as a
SCRATCH_DEV, where "device type is unknown" failure sees the/dev/mapper/$vgname-$lvnamenode never appear.Signed-off-by David Disseldorp <[email protected]>---
It might make more sense to add a default timeout to_udev_wait(), but that could also call udev{adm }settle.
One of the points of moving to _udev_wait was to explicitly wait for the specific device to appear or disappear, indicating that the previous admin operations have completed. 'udevadm settle' does not do that - it waits for the global queue of events to drain and dev config failure doesn't generate udev events.
Hence tests that silently fail dm/lvm device setup will continue to run on something they shouldn't have, and nobody will realise that the LVM setup did not run correctly.
OTOH, _udev_wait() will hang in that situation because it's waiting for the device node to appear (or disappear) so it can be immediately used. We don't need udev to finish draining queues - we only need to wait for the device node to appear and the test is good to go.
This behaviour provides obvious failures when device setup/teardown fails in some way. This should not happen, and when it does _udev_wait hanging forces that failure to be triaged and fixed immediately....
If we add a timeout to _udev_wait(), then we're back to the old behaviour where device config failures get ignored and the test runs incorrectly. Except now it is worse because we have to wait a timeout before the test is then run....
Quoted 11 lines
tests/f2fs/008 | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)tests/f2fs/008 +0
-0@@ -35,9 +35,12 @@ _cleanup() rm -f $tmp.* }
tests/f2fs/008 _cleanup() · line 43
Quoted 10 lines
tests/f2fs/008 +6
-3-$LVM_PROG pvcreate -f $SCRATCH_DEV >>$seqres.full 2>&1 -$LVM_PROG vgcreate -f $vgname $SCRATCH_DEV >>$seqres.full 2>&1 -$LVM_PROG lvcreate -y -L 1024m -n $lvname $vgname >>$seqres.full 2>&1 +$LVM_PROG pvcreate -f $SCRATCH_DEV >>$seqres.full 2>&1 \ + || _notrun "unabled to create LVM physical volume" +$LVM_PROG vgcreate -f $vgname $SCRATCH_DEV >>$seqres.full 2>&1 \ + || _notrun "unabled to create LVM volume group" +$LVM_PROG lvcreate -y -L 1024m -n $lvname $vgname >>$seqres.full 2>&1 \ + || _notrun "unabled to create LVM logical volume"
These should be _fail calls, IMO.
AFAICT, there is no reason for LVM config to fail on a valid block device. If there is a block device that doesn't support LVM, then we've got lots of tests that are going to fail the same way and we should have a _require_scratch_lvm() to notrun those tests for that type of block device (or fix the block device...).
-Dave.
Signature · 2 lines
Dave Chinner [email protected]
Re: [PATCH] avoid endless udevadm wait
[digging up an old thread]
Hi Dave,
Quoted 32 lines · Dave Chinner
Quoted 9 lines · David Disseldorp
"udevadm wait <dev>" without a --timeout=SECONDS parameter will wait endlessly. Endless wait can be triggered in f2fs/008 by e.g. using a zram device as a
SCRATCH_DEV, where "device type is unknown" failure sees the/dev/mapper/$vgname-$lvnamenode never appear.Signed-off-by David Disseldorp <[email protected]>---
It might make more sense to add a default timeout to_udev_wait(), but that could also call udev{adm }settle.One of the points of moving to
_udev_waitwas to explicitly wait for the specific device to appear or disappear, indicating that the previous admin operations have completed. 'udevadm settle' does not do that - it waits for the global queue of events to drain and dev config failure doesn't generate udev events.Hence tests that silently fail dm/lvm device setup will continue to run on something they shouldn't have, and nobody will realise that the LVM setup did not run correctly.
OTOH,
_udev_wait()will hang in that situation because it's waiting for the device node to appear (or disappear) so it can be immediately used. We don't need udev to finish draining queues - we only need to wait for the device node to appear and the test is good to go.This behaviour provides obvious failures when device setup/teardown fails in some way. This should not happen, and when it does
_udev_waithanging forces that failure to be triaged and fixed immediately....
Quoted 4 lines
If we add a timeout to
_udev_wait(), then we're back to the old behaviour where device config failures get ignored and the test runs incorrectly. Except now it is worse because we have to wait a timeout before the test is then run....
I've revisited this, as I manage to break udevd once a year or so in my minimal initramfs-based xfstests env. I find a timeout with golden-output breaking error message much easier to debug than an endless loop. Please see: