Description of problem: When running the attached test case on an ext3 file system eventually one of the processes using direct I/O (O_DIRECT) will fail with EIO. This has been reported to occur during for e.g. database load operations. This only occurs on kernels that include the patch: linux-2.6-fs-jbd-wait-for-t_sync_datalist-buf-to-complete.patch Version-Release number of selected component (if applicable): 2.6.18-53.1.13 onwards How reproducible: 100% Steps to Reproduce: 1. Compile the attached testcase with: $ gcc -Wall -D_GNU_SOURCE -o testcase testcase.c 3. Create a testfile: dd if=/dev/zero of=testfile bs=64k count=1000 3. Run multiple copies of the test in parallel with half using direct I/O and half using buffered I/O, e.g.: # ./testcase & ./testcase -d & ./testcase & ./testcase -d & ./testcase & ./testcase -d & ./testcase & ./testcase -d Actual results: [1] 18481 [2] 18482 [3] 18483 [4] 18484 [5] 18485 [6] 18486 [7] 18487 write failed: Input/output error Expected results: Test runs indefinitely without error Additional info: Several upstream threads discussing this: http://lkml.org/lkml/2008/5/1/160 http://lkml.org/lkml/2008/5/12/193
Created attachment 305460 [details] testcase to trigger O_DIRECT EIO problem
Mirroring events from IT This event sent from IssueTracker by balkov issue 172641
IT is refusing to mirror even when done manually... ----- Additional Comments From mranweil.com (prefers email at mjr.com) 2008-05-27 13:30 EDT ------- The testcase ran fine over the long weekend with the patch version 7. Elmar - it fix for you, too?
Created attachment 308846 [details] Patch correcting jbd races This is the final version of the patch pushed upstream by IBM. Now in -mm & expected to be merged in 2.6.26.
This request was evaluated by Red Hat Product Management for inclusion, but this component is not scheduled to be updated in the current Red Hat Enterprise Linux release. If you would like this request to be reviewed for the next minor release, ask your support representative to set the next rhel-x.y flag to "?".
Posted for review - pending PM ack based on Joe K.'s request http://post-office.corp.redhat.com/archives/rhkernel-list/2008-August/msg00097.html
This request was evaluated by Red Hat Product Management for inclusion in a Red Hat Enterprise Linux maintenance release. Product Management has requested further review of this request by Red Hat Engineering, for potential inclusion in a Red Hat Enterprise Linux Update release for currently deployed products. This request is not yet committed for inclusion in an Update release.
in kernel-2.6.18-104.el5 You can download this test kernel from http://people.redhat.com/dzickus/el5
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2009-0225.html