I heard 3.2.1 MU1 has been flagged with an issue anyone know

[hdtvguy]

We were going to do 3.2.1 MU1 and were tald it has a warning that was just issued, but no further details were released. We cancelled deployment until we can get details.

[Schmoog]

I'm on 3.2.1 MU1 and have been running with no issues.

There was a critical patch released for an issue with region moves on a TDVV (Thin Dedupe VV), but they issued a patch (P04) that fixed the issue. HP contacted me about the critical patch just after thanksgiving, and I had them install it over the weekend (Saturday and sunday after thanksgiving). No node reboot required, completely non-disruptive.

That being said, since we don't have SSD, and don't use thin dedupe, there were no worries for us. I just had them do it so I didn't have to think about it.

[RitonLaBevue]

Well, P04 is the key to MU1.

It's about dedup and moving region/chunklets of VVs in vvset setted for dedup and... data corruption ^^

So, no dedup in MU1 without P04.

[Schmoog]

Well, P04 is the key to MU1.

It's about dedup and moving region/chunklets of VVs in vvset setted for dedup and... data corruption ^^

So, no dedup in MU1 without P04.

Spot on. Here's the link to the customer bulletin:

http://h20566.www2.hp.com/hpsc/doc/public/display?docId=c04514111&jumpid=em_alerts_us-us_Nov14_xbu_all_all_2286082_201212_storage_critical_010_9700

[hdtvguy]

OK then I guess I will reschedule the install for after the new year.

[bajorgensen]

Also have one array on 3.2.1 MU1 since the release day without issue.
But HP came and installed P04 immediatly when it was released.

[Patrick]

Did a couple of upgrades from 3.1.3 MU1 to 3.2.1 MU1 with P04 and notice a completely different issue with iSCSI.

shownet -d is some issues on the iSCSI interfaces

Code: Select all

Detailed Tcp stats on node 0
RtoAlgorithm:                    1  RtoMin:                          200
RtoMax:                     120000  MaxConn:                          -1
ActiveOpens:                  6717  PassiveOpens:                   6768
AttemptFails:                   33  EstabResets:                      97
CurrEstab:                      89  InSegs:                   1111778630
OutSegs:                 496370425  RetransSegs:                  389274
InErrs:                          0  OutRsts:                          40

Detailed Tcp stats on node 1
RtoAlgorithm:                    1  RtoMin:                          200
RtoMax:                     120000  MaxConn:                          -1
ActiveOpens:                 39513  PassiveOpens:                  77477
AttemptFails:                  164  EstabResets:                   41280
CurrEstab:                      99  InSegs:                   1105276422
OutSegs:                 501448361  RetransSegs:                 1165450
InErrs:                          0  OutRsts:                        6799

10 seconds later

Code: Select all

Detailed Tcp stats on node 0
RtoAlgorithm:                    1  RtoMin:                          200
RtoMax:                     120000  MaxConn:                          -1
ActiveOpens:                  6717  PassiveOpens:                   6768
AttemptFails:                   33  EstabResets:                      97
CurrEstab:                      89  InSegs:                   1111782268
OutSegs:                 496375134  RetransSegs:                  389276
InErrs:                          0  OutRsts:                          40

Detailed Tcp stats on node 1
RtoAlgorithm:                    1  RtoMin:                          200
RtoMax:                     120000  MaxConn:                          -1
ActiveOpens:                 39513  PassiveOpens:                  77477
AttemptFails:                  164  EstabResets:                   41280
CurrEstab:                      99  InSegs:                   1105280815
OutSegs:                 501452729  RetransSegs:                 1165505
InErrs:                          0  OutRsts:                        6799

The above is from a 2-node 7400, also seeing similar issues on a pair of 7200's though the counts are less (much smaller configs).

Just ran across the issue a few hours ago when the VM farm disconnected/reconnected half of the exported LUNs....definitely something is up.

[Schmoog]

Fwiw we have had precisely zero issues with 3.2.1 mu1, but we run fc only (blech. The thought of iscsi makes me queasy)