December 2006 - Xymon - Xymon mailman web

Strange phenomena hobbit 4.2.0
by peter.welter＠gmail.com 27 Dec '06

27 Dec '06

In hobbit-clients.cfg the following is defined: HOST=somehost DIR /dumpdisk/backup/archivetmp In client-local.cfg [somehost] dir:/dumpdisk/backup/archivetmp In this configuration everything works just fine. Al defined 15 disk-partitions show up at the disk-test of somehost. However... When the TRACK-keyword is added in the hobbit-clients.cfg to the DIR-line above and the file filesizes.,dumpdisk,backup,archivetmp.rrd shows up in /var/lib/hobbit/rrd/somehost/ then the graphs that would normally be shown at the disk-test, do not work properly anymore! In fact, the first 5 partitions are not shown anymore. Removing the TRACK-keyword and re-moving file filesizes.,dumpdisk,backup,archivetmp.rrd makes the disk-graphs working again (yes ALL rrd-files are still updated). Regards, Peter

2 1

Error compiling todays (26/12) snapshot
by lars.ebeling＠leopg9.no-ip.org 27 Dec '06

27 Dec '06

Hi I have stll problems with compiling on HP-UX 11.11, First time I wrote about this was 30/11. ´ gcc -g -O -D_REENTRANT -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -DHPUX -I. -I /home/hobbit/snapshot/include -I/usr/local/include -c -o hobbitd_locator.o hobbi td_locator.c hobbitd_locator.c: In function `main': hobbitd_locator.c:746: error: `MSG_WAITALL' undeclared (first use in this functi on) hobbitd_locator.c:746: error: (Each undeclared identifier is reported only once hobbitd_locator.c:746: error: for each function it appears in.) hobbitd_locator.c:746: warning: passing arg 6 of `recvfrom' from incompatible po inter type hobbitd_locator.c:761: error: `MSG_DONTWAIT' undeclared (first use in this funct ion) make[1]: *** [hobbitd_locator.o] Error 1 make[1]: Leaving directory `/home/hobbit/snapshot/hobbitd' make: *** [hobbitd-build] Error 2 Regards Lars Ebeling http://leopg9.no-ip.org Hobbithobbyist "It is better to keep your mouth shut and appear stupid than to open it and remove all doubt." -- Mark Twain

3 2

Using mysql instead of filesystem
by jbeltram＠int.inetu.net 27 Dec '06

27 Dec '06

Is there any though about moving away from the flat filesystem for histlogs and logs, and go with a db solution like mysql? Thanks, Jason Beltrame

1 0

Todays snapshot (27/12)
by lars.ebeling＠leopg9.no-ip.org 27 Dec '06

27 Dec '06

I know that hobbitd_locator.c is work in progress but f.y.i gcc -g -O -D_REENTRANT -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -DHPUX -I. -I /home/hobbit/snapshot/include -I/usr/local/include -c -o hobbitd_locator.o hobbi td_locator.c hobbitd_locator.c: In function `recalc_current': hobbitd_locator.c:91: error: `INT_MAX' undeclared (first use in this function) hobbitd_locator.c:91: error: (Each undeclared identifier is reported only once hobbitd_locator.c:91: error: for each function it appears in.) hobbitd_locator.c: In function `main': hobbitd_locator.c:766: warning: passing arg 6 of `recvfrom' from incompatible po inter type make[1]: *** [hobbitd_locator.o] Error 1 make[1]: Leaving directory `/home/hobbit/snapshot/hobbitd' make: *** [hobbitd-build] Error 2 Regards Lars Ebeling http://leopg9.no-ip.org Hobbithobbyist "It is better to keep your mouth shut and appear stupid than to open it and remove all doubt." -- Mark Twain

2 3

[hobbit] New Hobbit "admin" Scripts
by michael.nemeth＠lmco.com 27 Dec '06

27 Dec '06

OK I got it to work. Changed : this_getQuery() { # RETURN=`echo "$1" | $GREP -oE "(^|[?&])$2=[^&]+" | $SED "s/%0D//g"| $SED "s/%0A/\n/g" | $CUT -f 2 -d "=" | $SED "s/%3D/=/g"` RETURN=`echo "$1" | awk -F\= '{print $2}' | $SED "s/%3D/=/g"` } What are you trying to do with all the grep and sed? BTW is almost always better (more portable?) to use awk -F rather than cut -f cut is "white space challenged " on many systems Also it Bash level dependent! Works: GNU bash, version 2.05.0(1)-release (sparc-sun-solaris2.8) Does work:GNU bash, version 2.03.0(1)-release (sparc-sun-solaris) -- +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ | _p_ Mike Nemeth | ___| |_____ email(w) michael.nemeth at lmco.com Work: 856 359-1425 |><___________) | Home Page:http://www.geocities.com/mjnemeth/ | Work Page:http://scorpion/~nemethm/ | Work Page:http://ortfsb/~mnemeth/ |++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

3 2

Double and missing alerts
by scott.brehm＠l-3com.com 27 Dec '06

27 Dec '06

I've been getting double alert mails for one of my servers, and no alerts for others. I'm wondering if the problem is related to my using the UNMATCHED rule. It's not listed on the man page for hobbit-alerts.cfg so I probably copied it from my earlier Big Brother config without thinking. Is there an UNMATCHED equivalent in Hobbit? Here's the current rules. I'm getting double e-mails for all tests on atscm-setnas2.set.local. I'm also getting no alerts for the http test on two servers that don't match the rules here. I didn't realize that until I followed the info link for them and saw No Alerts Defined in the alerting table. HOST=* SERVICE=conn MAIL scott.brehm at x.com COLOR=red HOST=%atscm MAIL scott.brehm at x.com EXSERVICE=conn COLOR=red HOST=%atscs MAIL scott.brehm at x.com EXSERVICE=conn DURATION>10 COLOR=red MAIL phil.lane at x.com DURATION>10 COLOR=red HOST=%atssl MAIL scott.brehm at x.com EXSERVICE=conn DURATION>10 COLOR=red MAIL stephanie.greer at x.com DURATION>10 COLOR=red HOST=%atswv MAIL scott.brehm at x.com EXSERVICE=conn DURATION>10 COLOR=red MAIL stephanie.greer at x.com DURATION>10 COLOR=red HOST=atscmmail1.ats.ad.local SERVICE=pop3 MAIL phil.lane at x.com COLOR=red HOST=setnas1.set.local MAIL scott.brehm at x.com EXSERVICE=conn COLOR=red MAIL charles.negus at x.com COLOR=red HOST=atscm-setnas2.set.local MAIL charles.negus at x.com COLOR=red UNMATCHED MAIL scott.brehm at x.com COLOR=red Scott Brehm IT Manager L-3 Communications - Applied Technologies 3394 Carmel Mountain Rd, San Diego CA, 92121 858-720-4035

2 1

AIX 51 ML07 bug
by Jim_Horwath＠glic.com 26 Dec '06

26 Dec '06

I have found a bug with AIX ML07 (not with the Hobbit code) on the operating system side. In the code hobbitclient-aix.sh, the line: netstat -an | grep "^tcp" gets caught in an infinite loop grep'ing a MAC. The program continues until the file system fills or until I am alerted for a purple condition and I correct the condition. I modified the line to this: netstat -an | grep "^tcp" | tail -5000 and it seems to fix it. I haven't heard anyone complain of this condition. It appears to happen only on AIX 5.1 ML07. Regards, Jim ----------------------------------------- This message, and any attachments to it, may contain information that is privileged, confidential, and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, you are notified that any use, dissemination, distribution, copying, or communication of this message is strictly prohibited. If you have received this message in error, please notify the sender immediately by return e-mail and delete the message and any attachments. Thank you.

1 0

hobbitfetch loop?
by dan.mcdonald＠austinenergy.com 22 Dec '06

22 Dec '06

I noticed that my hobbitfetch process had hung. It was consuming all available CPU. I ran strace for a while and saw no activity, so I tried killing the process, and saw this... [root at ldap hobbit]# strace -p 20466 Process 20466 attached - interrupt to quit --- SIGTERM (Terminated) @ 0 (0) --- time(NULL) = 1166824795 open("/etc/localtime", O_RDONLY) = 4 fstat64(4, {st_mode=S_IFREG|0644, st_size=1279, ...}) = 0 fstat64(4, {st_mode=S_IFREG|0644, st_size=1279, ...}) = 0 mmap2(NULL, 131072, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb7dd5000 read(4, "TZif\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\5\0\0\0\5\0"..., 131072) = 1279 close(4) = 0 munmap(0xb7dd5000, 131072) = 0 stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=1279, ...}) = 0 write(2, "2006-12-22 15:59:55 ", 20) = 20 write(2, "Caught TERM signal, terminating\n", 32) = 32 sigreturn() = ? (mask now []) --- SIGTERM (Terminated) @ 0 (0) --- time(NULL) = 1166824897 stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=1279, ...}) = 0 write(2, "2006-12-22 16:01:37 ", 20) = 20 write(2, "Caught TERM signal, terminating\n", 32) = 32 sigreturn() = ? (mask now []) +++ killed by SIGKILL +++ Process 20466 detached Any ideas on where to look? hobbitfetch.log is completely barren. -- Daniel J McDonald, CCIE # 2495, CISSP # 78281, CNX Austin Energy http://www.austinenergy.com

1 0

Missing primary oid data - DEVMON
by mprice＠hst.nasa.gov 22 Dec '06

22 Dec '06

Hello, When I am using DEVMON, somtimes my cisco switches get overloaded and wont return the oid data correctly. When this happens I check the hobbit web page and it says : Missing primary oid data This alarms "RED", I dont want to alarm on this, how can I fix this??? I just want it to alarm "clear" ... I dont want to totally ignore these alarms, but I dont want them to alarm red. I have tried using the "_AUTOMATCH_" field in the thresholds file and everything... any ideas??? thanks, michael -- Michael A. Price Performance Network Engineering NASA/GSFC Code 440.8/LMB Greenbelt, Maryland 20770 Phone: 240-684-1356 Cell: 410-507-7476 e-mail: mprice at hst.nasa.gov ----------------------------------------------

1 0

Hobbit for very large server installation
by grsamy＠yahoo.com 22 Dec '06

22 Dec '06

I am evaluating Hobbit for monitoring a network infrastructure with 3000+ servers. Do you have any hobbit screen shots for such large implementations? How 'Status "at a glance" ' can be achieved with hobbit for large number of servers? I am not new to bigbrother or Hobbit, but my maximum install base so far is only about 200 servers. Thanks Ram __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1