March 2008 - Xymon - Xymon mailman web

Alert Script page.log contains many: Argument list too long
by s_aiello＠comcast.net 26 Mar '08

26 Mar '08

Hello All, I was implementing a new alert script. After doing so I was checking my logs to make sure all was well. I noticed the server/logs/page.log had many: "Could not launch paging script /hobbit/script.sh: Argument list too long". So I thought there was something wrong with my script. So I changed my script to something very simple to debug and came up with nothing. So I started to dig into the code and found in hobbitd/do_alert.c: /* The child starts the script */ execlp(recip->scriptname, recip->scriptname, NULL); errprintf("Could not launch paging script %s: %s\n", recip->scriptname, strerror(errno)); exit(0); Searched for the execlp function and found it nowhere in the Hobbit code.. so it must be system level. Once I started to google for execlp + "Argument list too long", this seem to be quite a common problem. A snippet from one posting I found was: and the explanation: in the linux kernel is the function execve(), which is how all of the other exec() functions (execl, execlp, execle, etc.) are actually implemented. the way it works is by creating a 128K buffer at the top end of the memory space and copying the command line and environment for the new process into this space. it then loads the new program into memory, sets its argv and envp pointers, and jumps to its entry point. there's a lot more to it than that, but the point is that there is a 128K buffer which is the only thing "held" from the parent process to the child. the "Argument list too long" error message is actually the kernel's E2BIG error code, returned when the execve() is not able to fit the supplied argument list and environment into the 128K buffer. Hobbit server is 4.2.0 + allinone on a RHEL 4 update 4. So this is over my head, I know it isn't my script. My the status message is too big for the execlp buffer ? not sure if there is a way for me to increase that buffer, yet. Thanks, ~Steve

2 2

hobbit_snmpcollect configuration question
by sberkley＠mcclatchyinteractive.com 26 Mar '08

26 Mar '08

I'm trying to test out hobbit_snmpcollect and I'm not seeing new columns for the host in question or new rrd's in the data directory. I am seeing the full debug output in the logfile specified in hobbitlaunch.cfg, just nothing in hobbit. Do I need to make changes to bb-hosts as well? Here's what I've added to my configs: ---------------------------------------------------------- hobbitlaunch.cfg: [hobbit_snmpcollect] ENVFILE /home/hobbit/server/etc/hobbitserver.cfg NEEDS hobbitd CMD hobbit_snmpcollect --debug LOGFILE $BBSERVERLOGS/hobbit_snmpcollect.log INTERVAL 5m ---------------------------------------------------------- hobbit-snmpmibs.cfg: stock version (copied from the build directory) ---------------------------------------------------------- hobbit-snmphosts.cfg: [hostname.example.com] version=2 community=public systemmib ifmib ---------------------------------------------------------- Log message looks typically like this: ...(lots of successful-looking parsing messages for hobbit_snmpmibs.cfg)... 2008-03-26 12:41:54 Opening file /home/hobbit/server/etc/hobbit-snmphosts.cfg 2008-03-26 12:41:54 No more oids left 2008-03-26 12:41:54 Finished host hostname.example.com 2008-03-26 12:41:54 Transport setup is: 2008-03-26 12:41:54 bbdportnumber = 1984 2008-03-26 12:41:54 bbdispproxyhost = NONE 2008-03-26 12:41:54 bbdispproxyport = 0 2008-03-26 12:41:54 Recipient listed as '10.0.0.1' 2008-03-26 12:41:54 Standard BB protocol on port 1984 2008-03-26 12:41:54 Will connect to address 10.0.0.1 port 1984 2008-03-26 12:41:54 Connect status is 0 2008-03-26 12:41:54 Sent 2951 bytes 2008-03-26 12:41:54 Closing connection ---------------------------------------------------------- Nothing is coming in on the status channel. Should systemmib and ifmib be showing up as new columns for that host? -- S i m e o n B e r k l e y Systems Engineer Mcclatchy Interactive

1 0

Class designation Problem
by Robert-Manocchia＠IDEXX.com 26 Mar '08

26 Mar '08

I have the following alerts set up in hobbit-alerts.cfg: $DEVHOSTS=%(host1|host2) HOST=$DEVHOSTS SERVICE=disk MAIL me at there.com I have the following in hobbit-clients.cfg: CLASS=hpuxdev DISK * 96 98 On the hobbit page I'm still seeing a yellow alert for a filesystem on host1 at 92% stating that it has reached the warning level of 90%. I also started the hobbit client on host1 as follows ./runclient.sh -class=hpuxdev restart My question is with the above scenario and config why is host1 still sending yellow alerts with the default threshold of 90% instead of 96% as configured in the hobbit-clients.cfg file.. Thanks in advance.

1 0

Customizing the Alerts mails sent my hobbit
by arun.thangappandi＠ge.com 26 Mar '08

26 Mar '08

Hi, Can any body help me; how to customize the hobbit alert mails so that it shows the IP of the server which we monitor instead of showing the hostname in it. Thanks & Regards, Arunkumar

4 6

hobbit on freebsd
by daniel.rapp＠incabus.com 26 Mar '08

26 Mar '08

Hi i have a server with freebsd (FreeBSD xxx.xxx.xxx 6.2-RELEASE FreeBSD 6.2-RELEASE #0: Mon Jun 11 15:09:22 UTC 2007 root@:/usr/obj/usr/src/sys/SMP amd64) I have two problems, if anybody have any pointers that would be great.. Problem nr1) [root at xxx /usr/local/bb/client/logs]# tail hobbitclient.log Abort trap (core dumped) 2008-03-25 10:36:26 Unknown token 'r' ignored at line 219 2008-03-25 10:36:26 xstrdup: Cannot dup NULL string [root at xxx /usr/local/bb/client/bin]# tail /var/log/messages Mar 25 10:36:26 xxx kernel: pid 6496 (hobbitd_client), uid 1002: exited on signal 6 (core dumped) Mar 25 10:41:26 xxx kernel: pid 6535 (hobbitd_client), uid 1002: exited on signal 6 (core dumped) Mar 25 10:46:26 xxx kernel: pid 6593 (hobbitd_client), uid 1002: exited on signal 6 (core dumped) Problem nr2) from the bb server webpage: " purple Tue Mar 25 01:22:59 CET 2008 - Memory CRITICAL Memory Used Total Percentage red Physical 18446713700627775617M 1013M18446741075379662574% green Swap 0M 2048M 0% " That dosent seem correct.. [root at xxx /usr/local/bb/client/bin]# ./freebsd-meminfo Total:1013 Free:1924086628408 [root at xxx /usr/local/bb/client/bin]# vmstat procs memory page disks faults cpu r b w avm fre flt re pi po fr sr ad0 ad4 in sy cs us sy id 0 1 0 1530420 57368 40 1 0 0 194 70 0 0 659 1558 2068 1 2 97 It seems like there is something thats wrong when it comes to the mem info parsing.. Mvh Daniel Rapp

2 2

problem with dbcheck.pl
by bertrand.orcibal＠gpn.fr 26 Mar '08

26 Mar '08

Hello I want to install dbcheck.pl and there are lots of dependencies witch another PERL modules. How can i install all in one more time ? Bests regards.

2 1

Current snapshot compilation problem
by alevin＠reveredata.com 25 Mar '08

25 Mar '08

Hi, I've got a compilation problem with current snapshot: ... hobbit_snmpcollect.c:1068: error: 'PATH_MAX' undeclared (first use in this function) ... Adding: #include <limits.h> To the hobbit_snmpcollect.c solved it OS - FreeBSD 7.0 __________________________________ Alexandre Levin

7 14

hobbit email notifications with nullmailer
by waa-hobbitml＠revpol.com 25 Mar '08

25 Mar '08

I have a system (Gentoo) running hobbit. I have installed the 4.20 with the all-in-one patch. The system is using nullmailer as my preferred MTA. It is a very small, minimal qmail-like MTA run under daemontools. /usr/sbin/sendmail is the 'wrapper' for email submissions, but it does not accept a subject on the command line (eg: -s "this is the subject") the way the sendmail does: --[snip]-- # sendmail -h sendmail: option -h requires a value. usage: sendmail [flags] [recipients] <message Nullmailer sendmail emulator -B Ignored -b Ignored -C Ignored -d Ignored -F Set the full name of the sender -f Set the envelope sender address -h Ignored -i Ignored -L Ignored -m Ignored -N Ignored -n Ignored -O Ignored -o Ignored -p Ignored -q Ignored -R Ignored -r An alternate and obsolete form of the -f flag -t Read message for recipients -U Ignored -V Ignored -v Ignored -X Ignored -h, --help Display this help and exit --[snip]-- It requires that the email message being sent to (through?) it contains a "Subject:" line in the header. Without me changing to another MTA, is there a way to tell hobbit that this is what I need? Changing would be OK if that was the only way forme to get hobbit to send me notifications. Thanks! -- Bill Arlofski Reverse Polarity, LLC

4 5

hobbit email notifications with nullmailer
by waa-vai＠revpol.com 25 Mar '08

25 Mar '08

I have a system (Gentoo) running hobbit. I have installed the 4.20 with the all-in-one patch. The system is using nullmailer as my preferred MTA. It is a very small, minimal qmail-like MTA run under daemontools. /usr/sbin/sendmail is the 'wrapper' for email submissions, but it does not accept a subject on the command line (eg: -s "this is the subject") the way the sendmail does: --[snip]-- # sendmail -h sendmail: option -h requires a value. usage: sendmail [flags] [recipients] <message Nullmailer sendmail emulator -B Ignored -b Ignored -C Ignored -d Ignored -F Set the full name of the sender -f Set the envelope sender address -h Ignored -i Ignored -L Ignored -m Ignored -N Ignored -n Ignored -O Ignored -o Ignored -p Ignored -q Ignored -R Ignored -r An alternate and obsolete form of the -f flag -t Read message for recipients -U Ignored -V Ignored -v Ignored -X Ignored -h, --help Display this help and exit --[snip]-- It requires that the email message being sent to (through?) it contains a "Subject:" line in the header. Without me changing to another MTA, is there a way to tell hobbit that this is what I need? Changing would be OK if that was the only way forme to get hobbit to send me notifications. Thanks! -- Bill Arlofski Reverse Polarity, LLC

1 0

[hobbit] rrd-status.log messages
by waa-vai＠revpol.com 25 Mar '08

25 Mar '08

> Hi all, I am receiving the following messages in my rrd-status.log > file... > > Got over-size message, truncating at 131072 bytes (max 131072) > > Dropping (more) garbled data > Bad data in channel, skipping it. > > Everything seems to be working, but don't really know where to look to > see what host(s) this is coming from. > > Can someone tell me where I can change the max message size hobbit can > receive? Or RRD? > > Thanks, > Taylor Lewick Hi Taylor.. I see that no one has answered your question yet. I don't have an answer, but I just noticed that same thing in my rrd-status.log and found your post while looking for a reason/solution: 2008-03-25 13:06:53 Dropping (more) garbled data 2008-03-25 13:47:09 Got over-size message, truncating at 135167 bytes (max: 131072) 2008-03-25 13:47:09 Bad data in channel, skipping it 2008-03-25 13:47:09 Dropping (more) garbled data 2008-03-25 13:52:18 Got over-size message, truncating at 135167 bytes (max: 131072) 2008-03-25 13:52:18 Bad data in channel, skipping it Perhaps someone can point me in a direction too? -- Bill Arlofski Reverse Polarity, LLC

1 0