Oh yes indeed! Excellent! I even have my BEA column now! One quick question though, the bea column consists of only a bea heap utilization graph. Is there a number of other bea stats available? like the apache multiple graphs "LARRD:*,apache:apache1|apache2|apache3"? I have the two enterprise OID strings in Henrik's bea-snmpstats.sh ext file, but only one graph (heap) is displayed, and it's empty! I've done snmpwalks to the servers (to both OIDs) and get VAST amounts of data..
Rick
Spoke too soon. Now that a few polling periods have passed, the bea page is nothing but the truncated snmpwalk output from the BEA-WEBLOG-MIB and the graphs have disappeared, odd. in the bea.log file, "/home/hobbit/server/ext/bea-snmpstats.sh: line 27: /home/hobbit/server/bin/bb: Argument list too long" is being dumped at eat polling time. Any ideas where I went wrong?
Rick
On Thu, 2005-02-03 at 12:22, rwaegner wrote:
Oh yes indeed! Excellent! I even have my BEA column now! One quick question though, the bea column consists of only a bea heap utilization graph. Is there a number of other bea stats available? like the apache multiple graphs "LARRD:*,apache:apache1|apache2|apache3"? I have the two enterprise OID strings in Henrik's bea-snmpstats.sh ext file, but only one graph (heap) is displayed, and it's empty! I've done snmpwalks to the servers (to both OIDs) and get VAST amounts of data..
Rick
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
In <1107455777.32088.36.camel at rinux> rwaegner <rwaegner at questia.com> writes:
Spoke too soon. Now that a few polling periods have passed, the bea page is nothing but the truncated snmpwalk output from the BEA-WEBLOG-MIB and the graphs have disappeared, odd. in the bea.log file, "/home/hobbit/server/ext/bea-snmpstats.sh: line 27: /home/hobbit/server/bin/bb: Argument list too long" is being dumped at eat polling time. Any ideas where I went wrong?
How much data does the two snmpwalk commands in bea-snmpstats generate?
It's probably your servers that return more data than the current script can handle. The output of the snmpwalk commands are passed as a command-line parameter to the "bb" client tool, so if it exceeds something like 32 Kb (it varies between shell implementations), then your shell cannot handle it.
It would need some tweaking of the bea-snmpstats script to work around this, and you probably also need to recompile Hobbit with an increased MAXMSG setting.
Henrik
Lots of data. I have 1125 lines that total 132k. With ALL the cluster members responding to the management server that I am polling. The script is very simple, but the hobbitgraph.cfg for the bea stats is a little confusing. And what would need to change? Also, what needs to change for the MAXMSG setting?
Thanks
Rick
On Thu, 2005-02-03 at 15:52, Henrik Storner wrote:
In <1107455777.32088.36.camel at rinux> rwaegner <rwaegner at questia.com> writes:
Spoke too soon. Now that a few polling periods have passed, the bea page is nothing but the truncated snmpwalk output from the BEA-WEBLOG-MIB and the graphs have disappeared, odd. in the bea.log file, "/home/hobbit/server/ext/bea-snmpstats.sh: line 27: /home/hobbit/server/bin/bb: Argument list too long" is being dumped at eat polling time. Any ideas where I went wrong?
How much data does the two snmpwalk commands in bea-snmpstats generate?
It's probably your servers that return more data than the current script can handle. The output of the snmpwalk commands are passed as a command-line parameter to the "bb" client tool, so if it exceeds something like 32 Kb (it varies between shell implementations), then your shell cannot handle it.
It would need some tweaking of the bea-snmpstats script to work around this, and you probably also need to recompile Hobbit with an increased MAXMSG setting.
Henrik
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
One other thing, the data/rrd/beaserver directory has rrd files from the tests:
-rw-rw-r-- 1 hobbit hobbit 57516 Feb 3 16:52 bea.threads.questia_admin_server.weblogic.kernel.System.rrd -rw-rw-r-- 1 hobbit hobbit 57516 Feb 3 16:52 bea.threads.questia_admin_server.weblogic.kernel.Non-Blocking.rrd -rw-rw-r-- 1 hobbit hobbit 57516 Feb 3 16:52 bea.threads.questia_admin_server.weblogic.kernel.Default.rrd -rw-rw-r-- 1 hobbit hobbit 57516 Feb 3 16:52 bea.threads.questia_admin_server.weblogic.admin.RMI.rrd -rw-rw-r-- 1 hobbit hobbit 57516 Feb 3 16:52 bea.threads.questia_admin_server.weblogic.admin.HTTP.rrd -rw-rw-r-- 1 hobbit hobbit 171420 Feb 3 16:52 bea.memory.questia_admin_server.rrd
Appreciate any help you can provide!
Rick
On Thu, 2005-02-03 at 15:52, Henrik Storner wrote:
In <1107455777.32088.36.camel at rinux> rwaegner <rwaegner at questia.com> writes:
Spoke too soon. Now that a few polling periods have passed, the bea page is nothing but the truncated snmpwalk output from the BEA-WEBLOG-MIB and the graphs have disappeared, odd. in the bea.log file, "/home/hobbit/server/ext/bea-snmpstats.sh: line 27: /home/hobbit/server/bin/bb: Argument list too long" is being dumped at eat polling time. Any ideas where I went wrong?
How much data does the two snmpwalk commands in bea-snmpstats generate?
It's probably your servers that return more data than the current script can handle. The output of the snmpwalk commands are passed as a command-line parameter to the "bb" client tool, so if it exceeds something like 32 Kb (it varies between shell implementations), then your shell cannot handle it.
It would need some tweaking of the bea-snmpstats script to work around this, and you probably also need to recompile Hobbit with an increased MAXMSG setting.
Henrik
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
participants (2)
-
henrik＠hswn.dk
-
rwaegner＠questia.com