Log file monitoring

[jansonz]

Hi,

Is tit possible to capture several lines from log at once?

I'm trying to capture errors (with stack trace) in tomcat log files. They usually contain quite meaningful indications to problem I would like to capture those as well.

Janis

[zab_monkey]

Hi,

Sorry to ask but I want to clarify before offering an option;

Do you mean you want to search a number of lines at once and capture results, or;

Do you mean if you find a expression you are looking for, to capture in and the following, say, 10 lines or so?

Cheers,

JC

[jansonz]

Well,

Usually the error in Tomcat looks like this:
<Timestamp>ERROR java.io.FileNotFoundException: fred.txt
at java.io.FileInputStream.<init>(FileInputStream.jav a)
at java.io.FileInputStream.<init>(FileInputStream.jav a)
at ExTest.readMyFile(ExTest.java:19)
at ExTest.main(ExTest.java:7)

If I capture all the ERROR messages in the log with zabbix, it download only the line with "ERROR".

Do make some more specific triggers, I need to capture the following lines as well for Zabbix analysis.

Maybe anybody has some other solution in mind?

[alusvedejs]

is there an answer?

Hi!

looks like i am searching for similar solution

i am trying to monitor DB2 log file
and it also writes error message in several lines

to search for the error message in unix i use "grep -p Error"

v7-p7-prod 11:28:34 zabbix:/home/zabbix$ tail -100000 /home/db2inst1/sqllib/db2dump/db2diag.log | grep -p Error
2011-07-30-04.00.21.954691+180 I149135337A181 LEVEL: Error
PID:63242488 TID:49352 NODE:000 Title: SQLP_ALRCB
Dump File:/home/db2inst1/sqllib/db2dump/63242488.49352.000.dump.bin

2011-07-30-04.00.21.955063+180 I149135519A578 LEVEL: Error
PID : 63242488 TID : 49352 PROC : db2sysc 0
INSTANCE: db2inst1 NODE : 000 DB : RIDDB2
APPHDL : 0-57694 APPID: *LOCAL.db2inst1.110801091918
AUTHID : CDCINST1
EDUID : 49352 EDUNAME: db2agent (RIDDB2) 0
FUNCTION: DB2 UDB, data protection services, sqlp_AsyncLogReadAPI, probe:210
DATA #1 : String, 132 bytes
Error: sqlcode -2657, rc 262144, hflag2LsnReuse 0
action 1, startLSN 000000769C966801, endLSN 00000076D039E44A, logBufferSize 204800

2011-07-30-05.00.34.216699+180 I149146247A181 LEVEL: Error
PID:63242488 TID:44211 NODE:000 Title: SQLP_ALRCB
Dump File:/home/db2inst1/sqllib/db2dump/63242488.44211.000.dump.bin

2011-07-30-05.00.34.216921+180 I149146429A578 LEVEL: Error
PID : 63242488 TID : 44211 PROC : db2sysc 0
INSTANCE: db2inst1 NODE : 000 DB : RIDDB2
APPHDL : 0-760 APPID: *LOCAL.db2inst1.110801113231
AUTHID : CDCINST1
EDUID : 44211 EDUNAME: db2agent (RIDDB2) 0
FUNCTION: DB2 UDB, data protection services, sqlp_AsyncLogReadAPI, probe:210
DATA #1 : String, 132 bytes
Error: sqlcode -2657, rc 262144, hflag2LsnReuse 0
action 1, startLSN 000000769E4060AB, endLSN 00000076D089C5C7, logBufferSize 204800


but if i am not using "-p" switch for grep i get meaningless information:

v7-p7-prod 11:29:56 zabbix:/home/zabbix$ tail -100000 /home/db2inst1/sqllib/db2dump/db2diag.log | grep Error
2011-07-30-04.00.21.954691+180 I149135337A181 LEVEL: Error
2011-07-30-04.00.21.955063+180 I149135519A578 LEVEL: Error

and if i create zabbix (active) item:
log["/home/db2inst1/sqllib/db2dump/db2diag.log","Error","UTF-8",100]

i also get meaningless information that tells me "something is wrong there"
but is not giving me a clue shat exactly is happening.

how can i add "-P" flag for grep in zabbix (active) item ?