Bug 526259

This request is past the original due dates for feature requests for RHEL 5.5, but I'm wondering if Mike Christie already has another bugzilla queuing up this work. Mike?

I was going to update for 5.5, but I did not have a bugzilla for it yet.

Looks like you can use this one then. Thanks Mike!

Created attachment 363067 [details]
Remove unnecessary warning

Created attachment 363068 [details]
fix debug macros

Created attachment 363069 [details]
use scsi host number in dbg msgs

Created attachment 363070 [details]
Rename lport NONE state to lport DISABLED

Created attachment 363071 [details]
Dont login back if going down

Created attachment 363072 [details]
Flush all rports deferred work when link goes down

Created attachment 363073 [details]
Rename rport NONE to rport DELETE

Created attachment 363074 [details]
dont drop lock during rport logoff

Created attachment 363075 [details]
cancel rport retry timer when rport goes away

Created attachment 363076 [details]
fix misleading fc_exch debug statement

Created attachment 363077 [details]
split rport-priv and libfc-rport

Created attachment 363078 [details]
change rport_create internface

Created attachment 363079 [details]
fix rport priv macros

Created attachment 363080 [details]
scsi/fcp and rport/disc use different interfaces to rports

Created attachment 363081 [details]
elsct interface uses fcid

Created attachment 363082 [details]
detach libfc's rport from scsi transport's rport

Created attachment 363083 [details]
remove unused lookup all rports

Created attachment 363084 [details]
rearrange fc_rport_work deferred work

Created attachment 363085 [details]
rename rport CREATED to rport READY in prep for ADISC

Created attachment 363086 [details]
get rid of rogues

Created attachment 363087 [details]
fix rport event race between logo and ready

Created attachment 363088 [details]
remove event from disc, pass it directly

Created attachment 363089 [details]
remove unused disc delay

Created attachment 363090 [details]
print 0 for frame pointers and actual error numbers for other errors

Created attachment 363091 [details]
simplify lport-rport callback since only for dns rport

Created attachment 363092 [details]
move rport list to rport module, still under discovery lock

Created attachment 363093 [details]
tarball of all patches

Thanks Abhijeet.  Are these patches and the ones in 
519091 patches you wanted in 5.5 and in 5.4.z?

Mike - I think the confusion is around if libfc is going to be re-based or not. If it is, then these patches may not be even applicable, since they are against 5.4's libfc.

So, if 5.5 isn't going to be re-based, then I would assume these patches are requested for 5.5, else - Cisco will need to test an early 5.5. with the re-based libfc to see how it impacts fnic and/or regressions.

Am I on track here, Abhijeet?

That's right Andrius. 

1) We need these patches for 5.4-z. These are only libfc bug fixes and improvements to libfc disc/rport modules, and do not include any new features like FIP or NPIV.

2) If 5.5 libfc is rebased, Mike would automatically pick these patches and many other non-bug-fix patches for 5.5. from upstream libfc (like FIP support for instance)

3) If 5.5 is not rebased from upstream, then these patches would be required for 5.5 too.

thanks.

@Cisco,

We need to confirm that there is commitment to test 
for the resolution of this request during the RHEL 5.5 test
phase, if it is accepted into the release. 

Please post a confirmation before Oct 16th, 2009, 
including the contact information for testing engineers.

yes, this would get tested as part of fnic/libfc for 5.5 after the rebase is done from upstream/open-fcoe.

Hi Mike, how is this looking for 5.5?

~~ Attention Customers and Partners - RHEL 5.5 Beta is now available on RHN ~~

RHEL 5.5 Beta has been released! There should be a fix present in this 
release that addresses your request. Please test and report back results 
here, by March 3rd 2010 (2010-03-03) or sooner.

Upon successful verification of this request, post your results and update 
the Verified field in Bugzilla with the appropriate value.

If you encounter any issues while testing, please describe them and set 
this bug into NEED_INFO. If you encounter new defects or have additional 
patch(es) to request for inclusion, please clone this bug per each request
and escalate through your support representative.

@Cisco,

Please update this bugzilla with results from testing as soon as possible. Thank you.

NetApp cert suite (NATE) was run against the RHEL 5.5 driver in non-FIP mode on UCS. All tests passed fine.

Here's a short description of the tests from our Cert Engineer:

=======
NATE starts IO to each of the LUNs and proceeds to tune the I/O level so that it falls within NTAP guidelines (45-85% filer utilization).  Once tuned, NATE proceeds through a series of fault injection steps The faults are:
Take-give (5)
Forced take-give (5)
Filer panic (5)
Filer forced reboot (5)
Filer switch port block (50)
Filer wwpn switch port block (50)
FCP down (5)

The numbers in parentheses are the iteration count of the test.

During all of these tests, DT (Disk Test) is running I/O patterns on the disks attached to the OS on the blades (or virtual machines).  If DT were to see any data corruption, it would be immediately flagged as a failure.  The hosts are checked that they didn't reboot during the test, nor did they lose ethernet connectivity.  Also, the hosts are checked after each filer event to make sure the proper number of disk devices and paths are visible on the host.

When all totaled, NATE ran 1272 separate tests against the FC portion of Palo on RHEL-5.5.

=======

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-2010-0178.html