Locating Truncation Issues |
If you suspect truncation issues, search for TRUNCATION in the tsunami log.
Note: If you regularly see truncation in an hour or less, the DXi system is too small.
From the diagnostics file:
grep TRUNCATION scratch/collect/node1-collection/app-info/tsunami.log
From the system:
grep TRUNCATION /var/log/DXi/tsunami.log
EXAMPLE OUTPUT
WARN - 05/13/09-11:06:45 - dedupd dedup.cpp(1599) [/opt/DXi/dedupd]
start_trunc_event_handler() - Received SM_START_TRUNCATION_EVENT !!
WARN - 05/13/09-14:29:57 - dedupd dedup.cpp(1615) [/opt/DXi/dedupd]
stop_trunc_event_handler() - Received SM_STOP_TRUNCATION_EVENT !!
What does this example output tell you?
This example shows that the customer reached the Truncation level at 11:06:45 and came out of truncation at 14:29:57. Use this information when looking at ingest, blockpool, deduplication, and bpgc performance issues, and whem performing troubleshooting.
From the diagnostics file:
grep Truncate
scratch/collect/node1-collection/app-info/pse_snapshot/usr/adic/SRVCLOG/logs/adminLog.hist
From the system:
grep Truncate /usr/adic/SRVCLOG/logs/adminLog.hist
EXAMPLE OUTPUT
May 13 14:29:57 2009 102 1913258518 SpaceManagementDaemon <subject_line>Space Management: Cleared Immediate Truncate Mode</subject_line><email_body>The system is not currently truncating original files after de-duplication. The system freespace is under the truncation threshold.</email_body>
What does this example output tell you?
The message output pretty much says it all. The DXi system is no longer truncating original files, and has resumed normal disk space mode.
Right after the above admin alert, you will see:
SpaceManagementDaemon <subject_line>Space Management: Resumed Normal Disk Space Mode</subject_line><email_body>The system is not currently truncating original files after de-duplication. The system freespace is under the truncation threshold.</email_body>
Notes |
This page was generated by the BrainKeeper Enterprise Wiki, © 2018 |