I believe this occurred while running disk/disk_stress_ng_dm-1 from the certification test suite on the system in our lab called "seuss", which had been installed manually w/ d-i, using LVM w/ crypted home. (I dug that info out of LP: #1749040, from whence this bug was spawned).
Since this crash occurred in filesystem code - and we didn't see this crash in other configs - it's possible that an LVM crypt setup is relevant to the reproduction.
My suggestion for next steps would be to recreate this setup (same kernel, partitioning scheme, running the test from $HOME), and run the test several times (maybe overnight in a loop) to see if we can reproduce the same failure. Note that at least one other failure has been observed while running this test - what we're looking for is another panic in jbd code.
I believe this occurred while running disk/disk_ stress_ ng_dm-1 from the certification test suite on the system in our lab called "seuss", which had been installed manually w/ d-i, using LVM w/ crypted home. (I dug that info out of LP: #1749040, from whence this bug was spawned).
Since this crash occurred in filesystem code - and we didn't see this crash in other configs - it's possible that an LVM crypt setup is relevant to the reproduction.
My suggestion for next steps would be to recreate this setup (same kernel, partitioning scheme, running the test from $HOME), and run the test several times (maybe overnight in a loop) to see if we can reproduce the same failure. Note that at least one other failure has been observed while running this test - what we're looking for is another panic in jbd code.