Hide Forgot
Description of problem: There is no module bfa for Brocade FC adapter in rt-kernel. Kernel for RHEL 5.6 (2.6.18-238.12.1.el5) has bfa module. Version-Release number of selected component (if applicable): 2.6.33.9-rt31.64.el5rt How reproducible: Boot last version of kernel-rt. # modinfo bfa modinfo: could not find module bfa Expected results: Module bfa is present in kernel package ( modinfo bfa print actual module information Additional info: Recompiling kernel_rt with CONFIG_SCSI_BFA_FC=m fix this problem, but compiled bfa module older than one in 2.6.18...(RHEL5.6): 2.0.0 vs 2.1.2
Unfortunately problem was not fixed with recompiled kernel.
Backported the bfa driver from RHEL6.1 to MRG-2.0, kernel-rt-2.6.33.9-rt31.70.el6rt.
Sorry, but this version is not available on RHN and/or ftp.redhat.com. The latest version is kernel-rt-2.6.33.9-rt31.66.el6rt.
Those fixes will be part of the next Kernel Errata, to be released soon. Even though I don't have the exact date for the Errata release, I expect that to happen in a couple of weeks.
Thank you for clarifying situation.
Technical note added. If any revisions are required, please edit the "Technical Notes" field accordingly. All revisions will be proofread by the Engineering Content Services team. New Contents: Cause: the Brocade FC adapter driver was disabled on mrg-2.0 and and its version was older thatn the one in RHEL6. Consequence: No Brocade FC driver was shipped with mrg-2.0 GA kernel. Fix: upgrade the bfa driver in mrg-2.0 and enable it in the build process. Result: mrg-2.0 now ships the bfa driver.
I have a question about MRG 1.3. Do you plan to include Brocade FC adapter driver into MRG 1.3 kernel? As I understand, kernel-rt versions in 1.3 and 2.0 are the same (2.6.33-9-rt31 it this moment).
No plans to. MRG 1.3 is in maintenance mode now.
Verified that Brocade FC is enabled in kernel-rt-2.6.33.9-rt31.74.el6rt. Done some disk tests and stress tests in a multipath setup. Verified against kernel-rt-2.6.33.9-rt31.66.el6rt, which did not find the disks on the bfa adapter. One remark: Running aio-stress test can trigger a kind of lock. That was possible to get out of by boosting bfa_wq_* threads to SCHED_FIFO:1. This should probably be put into the release notes. For optimal disk performance, it might be needed to work more on such tweaks.
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/RHBA-2011-1192.html