Serious deadlock problem in 1.6.6

[bkowalewski]

We are getting flooded with this deadlock error........ Any ideas?

We are 1.6.6, a master and 1 proxy, MySQL, Solaris 10, Sun T2000. Around 1800 hosts, around 43,000 items, 91 new values/sec.

1168:20091203:071047 [Z3005] Query failed: [1213] Deadlock found when trying to get lock; try restarting transaction [update items set nextcheck=1259838619,prevvalue=lastvalue,lastvalue ='The operating system returned error 665(The requested operation could not be completed due to a file system limitation) to SQL Server during a write at offset 0x0000070533c000 in file \'S:\\WideOrbit\\Data\\WO_TRAFFIC.mdf:MSSQL_DBCC9\ '. Additional messages in ',lastclock=1259838439,lastlogsize=77575 where itemid=78303]
1162:20091203:071047 [Z3005] Query failed: [1213] Deadlock found when trying to get lock; try restarting transaction [update items set nextcheck=1259842811,prevvalue=lastvalue,lastvalue ='S:\\WideOrbit\\Data\\WO_TRAFFIC.mdf:MSSQL_DBCC9: Operating system error 665(The requested operation could not be completed due to a file system limitation) encountered.

[NOB]

Hi bkowalewski

We are getting flooded with this deadlock error........ Any ideas?

We are 1.6.6, a master and 1 proxy, MySQL, Solaris 10, Sun T2000. Around 1800 hosts, around 43,000 items, 91 new values/sec.

1168:20091203:071047 [Z3005] Query failed: [1213] Deadlock found when trying to get lock; try restarting transaction [update items set nextcheck=1259838619,prevvalue=lastvalue,lastvalue ='The operating system returned error 665(The requested operation could not be completed due to a file system limitation) to SQL Server during a write at offset 0x0000070533c000 in file \'S:\\WideOrbit\\Data\\WO_TRAFFIC.mdf:MSSQL_DBCC9\ '. Additional messages in ',lastclock=1259838439,lastlogsize=77575 where itemid=78303]
1162:20091203:071047 [Z3005] Query failed: [1213] Deadlock found when trying to get lock; try restarting transaction [update items set nextcheck=1259842811,prevvalue=lastvalue,lastvalue ='S:\\WideOrbit\\Data\\WO_TRAFFIC.mdf:MSSQL_DBCC9: Operating system error 665(The requested operation could not be completed due to a file system limitation) encountered.

AFAIR we did experience the same (or similar) problem on the same platform
using a Solaris 10 zone with disk on SAN storage.

We get rid of it by setting "StartDBSyncers=1" in the zabbix server configuration
file.
This never happens on a real modern platform like HP 380 G5
with "only" 8 cores, 4 GB of RAM and 15k SAS drives.
It's not the first case we experienced that the new 32-core Suns
are slow.

HTH,

Norbert.