+++ This bug was initially created as a clone of Bug #745676 +++ Description of problem: When a target sends a nop bnx2i might copy the target transfer tag incorrectly. The target then might drop the session causing IO to be failed. This was observed when connected to a Dell nx4 iSCSI target on the RHEL5.8 2.6.18-296.el5 kernel via bnx2i. Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: --- Additional comment from mchristi on 2011-10-12 21:45:04 EDT --- Exception notes: This is a pretty simple fix that has been upstream for a while, so it has been widely tested. It only affects broadcom bnx2i setups. And it occurs when using bnx2i with a popular target from Dell, so it affects many users. --- Additional comment from pm-rhel on 2011-10-12 21:50:40 EDT --- 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. --- Additional comment from mchristi on 2011-10-12 21:55:10 EDT --- QE, You would need bnx2i connected to something like a Equallogic target to test this. I am not sure if we have a setup like this. Broadcom will test that case. Our normal testing will test for regressions, so you should not have to do anything extra. --- Additional comment from eddie.wai on 2011-10-13 03:10:06 EDT --- Broadcom just verified the following with the 57810 NIC: - the RHEL6.2 SS1 inbox bnx2i driver does indeed show the session flap issue as triggered by ping timeouts which are caused by NOP reply failures. - the driver sent from Mike Christie, which is based on SS1 + the aforementioned TTT patch, no longer shows the session flap issue. --- Additional comment from arozansk on 2011-10-19 11:02:20 EDT --- Patch(es) available on kernel-2.6.32-210.el6 --- Additional comment from bdonahue on 2011-10-27 14:09:23 EDT --- We ran our tests, did a SanityOnly check and Broadcomm verified the fix (comment 4).
This is the upstream patch which will resolve the issue. Please incorporate this into the next release. Thanks. commit 610602f369b4c810c9df05e431abd38f38cb8e0d Author: Eddie Wai <eddie.wai> Date: Fri Aug 26 11:16:47 2011 -0700 [SCSI] bnx2i: Fixed the endian on TTT for NOP out transmission The iscsi_nopout task's TTT is defined as __be32 while the DMA memory to the chip is CPU specific. This creates a problem for unsolicited NOP-In responses where the TTT is not the RESERVED tag of 0xFFs. This patch adds a call to be32_to_cpu for the TTT specified. Signed-off-by: Eddie Wai <eddie.wai> Signed-off-by: James Bottomley <JBottomley>
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.
Patch(es) available in kernel-2.6.18-300.el5 You can download this test kernel (or newer) from http://people.redhat.com/jwilson/el5/ Detailed testing feedback is always welcomed. If you require guidance regarding testing, please ask the bug assignee.
Eddie, To test this would we want to log into a target that sends nops, like Equallogic, then not send any IO? The target would then send nops as pings to the initiator, right?
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHSA-2012-0150.html