| |||||||||||||
It's not about OracleIt's not about OraclePosted Mar 5, 2011 1:33 UTC (Sat) by marrusl (subscriber, #67123)In reply to: It's not about Oracle by wzzrd Parent article: Commitment to Open (Red Hat News)
Novell also sells support contracts for RHEL.
(Log in to post comments)
It's not about Oracle Posted Mar 5, 2011 14:28 UTC (Sat) by erwbgy (subscriber, #4104) [Link]
Correct. The supplier that the company I work for uses Novell because they are cheaper than Red Hat. From my side I haven't be too impressed with the results though.
It's not about Oracle Posted Mar 6, 2011 0:30 UTC (Sun) by csamuel (✭ supporter ✭, #2624) [Link]
Heh, as someone who has had bug reports in on the RHEL kernel with RH I wouldn't bet they would do much better. Would be nice if they considered packets being delivered on the wrong 10GigE interface something they should fix sooner rather than when 5.7 (eventually) comes out (reported prior to 5.6 release, but apparently too late to be considered)..
It's not about Oracle Posted Mar 6, 2011 4:56 UTC (Sun) by airlied (subscriber, #9104) [Link]
did you ask support for a hotfix or Z stream fix?
It's not about Oracle Posted Mar 6, 2011 7:18 UTC (Sun) by csamuel (✭ supporter ✭, #2624) [Link]
We are lucky enough to be able to work around the bug by bonding the interfaces together, tagging them and then passing both VLAN's down both interfaces so it no longer matters which interface packets are delivered on (plus we get some redundancy into the bargain). Others might not be so fortunate (and I've found at least one bugzilla entry [1] that indicates that a large US HPC lab has stumbled across the same issue that "upsets" their Lustre filesystem).
On the other hand our support contact at Red Hat seems to have trouble getting any info out of the RH kernel folks. We've determined ourselves that the fix is a newer version of the Mellanox driver and fed that back to them and I suspect that's what they are planning for RHEL 5.7. I hope.
|
|||||||||||||