Unsolved
This post is more than 5 years old
23 Posts
0
4641
Access to VMAX internal logs?
We recently encountered data corruption due to a broken SFP. Not sure on which end (VMAX or SAN switch) since both were replaced. There was no alert in SMC and even ESRS Call-Home was not triggered. After I opened a ticket the engineer could see the internal log of the VMAX and saw BB2D alerts.
We had luck since the error happened during a VMware installation so no data was lost. But now I am concerned that it could happen again and we have no way of getting alerted or at least to check ourselves if there are any errors. "symevent list" does not show anything related.
The remote support engineer and the CE on site had easy access to these log files. How can we see these log entries?
Thanks!
rmueller2
23 Posts
0
July 31st, 2012 05:00
I used both paths when I installed the servers. After I saw that there was an issue I disabled path A first. Issue was still there. Then I disabled path B and re-enabled path A and had no issues. Hence I knew, that path B was faulty. I included all those details in the ticket and the engineer could confirm that the errors were exactly on this port of path B.
Vipin VK
812 Posts
0
July 31st, 2012 05:00
Disabling one path didn't cause any issue, means it is working with single path active. How you tried swapping the paths ?
Have you reported this issue to the onsite CE/ remote engineer ?
Vipin VK
812 Posts
0
July 31st, 2012 05:00
Also, data corruption due to single SFP issue ?
sauravrohilla
859 Posts
0
July 31st, 2012 05:00
I am afraid you cant. Only EMC authorized person can get the access to these errors through INLINE commands.
rmueller2
23 Posts
0
July 31st, 2012 05:00
The ESXi 5.0 installation stopped at 34% every time for a bunch of servers. The error message was a timeout. I disabled one FC path and all was fine. Once I swapped the paths the error came back. I had no chance to see any error, I could just guess. After the SFPs were replaced I could install all servers with both FC-paths active.
Vipin VK
812 Posts
0
July 31st, 2012 09:00
ok,. So one path was down (SFP fault), you disabled the second and then started the issue. There the VMAX is innocent ..
And regarding the logs, as per my understanding those are for service purpose and are not to be accessed by customers.The same is explained by Sourabh here.
Vipin V.K
HankD1
20 Posts
0
July 31st, 2012 12:00
Did you try symevent?
C:\Program Files\EMC\SYMCLI\bin>symevent list
Symmetrix ID:
Time Zone : Central Daylight Time
Detection time Dir Src Category Severity Error Num
------------------------ ------ ---- ------------ ------------ ----------
Fri Dec 16 13:50:52 2011 FA-5H Symm Device(2E6C) Warning 0x003f
Access was attempted to a Not Ready device
Fri Dec 16 13:50:53 2011 FA-11H Symm Device(2E6C) Warning 0x003f
Access was attempted to a Not Ready device
Wed Jun 27 15:03:10 2012 FA-5F Symm Device(1ED8) Warning 0x003f
Access was attempted to a Not Ready device
Wed Jun 27 15:03:10 2012 FA-11F Symm Device(1ED8) Warning 0x003f
Access was attempted to a Not Ready device
Wed Jun 27 17:47:03 2012 FA-11G Symm Device(273A) Warning 0x003f
Access was attempted to a Not Ready device
rmueller2
23 Posts
0
July 31st, 2012 22:00
I tried symevent list as mentioned above but it did not show anything: "No Symmetrix Internal events were found to report on"
PedalHarder
465 Posts
0
August 1st, 2012 22:00
Not every single event that occurs in the array is reported via symevent. Nor is every event dialed home to EMC. There is a defined list of events that will get reported to the CLI. You can see a list of them with 'man symevent'.
In your case i expect you should have seen some error counters going up on the switch port (CRC events etc.). Worth looking on the switch when experiencing your type of symptoms.