February 2014 - Xymon - Xymon mailman web

canceling acknowledgements
by lebarber＠gmail.com 12 Mar '14

12 Mar '14

Is there some way to cancel an 'ack' after it has been issued? Thanks, Larry Barber

4 5

BBwin 0.13 and Mountpoints
by rower.master＠gmail.com 10 Mar '14

10 Mar '14

Hi All, I´m trying to fix the issue with the graphs creation for windows mountpoints and bbwin 0.13. The issue is that we only have graphs for the single drive letters and not for the mountpoints using the bbwin 0.13. Another problem is even if the graphs were working we could not have duplicated names for different drive letters. Filesystem 1K-blocks Used Avail Capacity Total Size Free Space Type Status Mount Point C 62811132 49966236 12844896 79% 59.90 GB 12.25 GB FIXED green N/A D 116043772 1100880 114942892 0% 110.67 GB 109.62 GB FIXED green N/A E 943848444 533077788 410770656 56% 900.12 GB 391.74 GB FIXED green N/A F 943848444 495800916 448047528 52% 900.12 GB 427.29 GB FIXED green N/A J 20970492 14764648 6205844 70% 20.00 GB 5.92 GB FIXED green N/A DadosLocal_Diversos_A 1237451772 928809832 308641940 75% 1.15 TB 294.34 GB FIXED green J:\DadosLocal_Diversos_A\ DadosLocal_Diversos_B 1782825980 1216555488 566270492 68% 1.66 TB 540.04 GB FIXED green J:\DadosLocal_Diversos_B\ Tempdb 104869884 54526796 50343088 51% 100.01 GB 48.01 GB FIXED green J:\Tempdb\ Log_Local 199235580 135337320 63898260 67% 190.01 GB 60.94 GB FIXED green J:\Log_Local\ I´ve already modified the winPSclient to consider the mountpoints and If I use this client I have all the graphs created without any problem. But I don´t want to continue changing or adding a server agent everytime a server administrator configure a mountpoint in a server. WinPSCLient output example where I have all graphs created: Mon 06 Jan 16:44:55 2014 - Filesystems OK Filesystem 1K-blocks Used Avail Capacity Mounted Summary(Total\Avail GB) C 52377596 39455096 12922500 75% /FIXED/C 49.95gb\12.32gb D 90254332 82510624 7743708 91% /FIXED/D 86.07gb\7.38gb F 10482684 91992 10390692 1% /FIXED/F 10.00gb\9.91gb F:\ORAARCHIVE 10482684 85608 10397076 1% /FIXED/F:\ORAARCHIVE 10.00gb\9.92gb F:\ORABACKUP01 10482684 1638116 8844568 16% /FIXED/F:\ORABACKUP01 10.00gb\8.43gb F:\ORADATA01 41939964 31552684 10387280 75% /FIXED/F:\ORADATA01 40.00gb\9.91gb I 10482684 84468 10398216 1% /FIXED/I 10.00gb\9.92gb I:\ORAARCHIVE 5239804 54772 5185032 1% /FIXED/I:\ORAARCHIVE 5.00gb\4.94gb I:\ORABACKUP01 52425724 93492 52332232 0% /FIXED/I:\ORABACKUP01 50.00gb\49.91gb I:\ORADATA01 209712124 98292 209613832 0% /FIXED/I:\ORADATA01 200.00gb\199.90gb J 10482684 96676 10386008 1% /FIXED/J 10.00gb\9.90gb J:\OraArchive 5239804 125628 5114176 2% /FIXED/J:\OraArchive 5.00gb\4.88gb J:\ORADATA01 207614972 136793092 70821880 66% /FIXED/J:\ORADATA01 198.00gb\67.54gb J:\ORATEMP 41939964 33645564 8294400 80% /FIXED/J:\ORATEMP 40.00gb\7.91gb I´ve tried some modifications on do_disk.c but without sucess. I´ve created a new dsystype called DT_MOUNTPOINT: *else if (strstr(msg, "Mount Point")) dsystype = DT_MOUNTPOINT;... if ( (dsystype == DT_MOUNTPOINT) && (*(columns[9])) && (strchr(columns[11], '/')) ) dsystype = DT_MOUNTPOINT;if ( (dsystype == DT_MOUNTPOINT) && (*(columns[9])) && (strchr(columns[11], ':')) ) dsystype = DT_MOUNTPOINT;* And here at the "case DT_MOUNTPOINT" is where the things are not good. I´ve tried to get the the column[11] but unfortunately we have backslashes. I´ve tried several ways, but I´m not good in c as I would like to and after many many tests I´m asking for a help or if you can put this on Xymon todo list. I´m using xymon 4.3.0-beta2 , sorry if this was already solved in the last version. I´ve checked the do_disk.c file on the last stable but couldn´t see this solution. *case DT_MOUNTPOINT: diskname = xstrdup(columns[11]); if (strstr(diskname, "N/A")) { diskname = xstrdup(columns[0]); } else { diskname = xstrdup(columns[11]); } /*sprintf(diskname, "/%s", columns[11]);*/ p = strchr(columns[4], '%'); if (p) *p = ' '; pused = atoi(columns[4]); aused = str2ll(columns[2], NULL); break;* I´ve tried to use the diskname as columns[0] but doesn´t work for the mountpoints too. The case DT_BBWin format that works for the winPSclient doesn´t work if I use the diskname as columns[0] or columns[11]. case DT_BBWIN: *diskname = xmalloc(strlen(columns[0])+2); sprintf(diskname, "/%s", columns[0]);* Thanks in advance, Mario

3 4

umlauts in xymon
by hj.beckers＠kreis-steinfurt.de 05 Mar '14

05 Mar '14

As the title says: If I enter a "ä" into the following directive "group-only apps|... |ikol|laemmkomm| ..." instead of the "ae" the column is ignored when xymon displays the page. "file" on linux characterizes the file as UTF-8, firefox page information shows the same char set. Coding the "ä" as a html entity makes no difference. Yours Hermann-Josef Beckers

2 3

nosslcert
by tawilcox＠fs.fed.us 04 Mar '14

04 Mar '14

I am just starting out with Xymon and have many questions. The current issue is that I want to suppress the sslcert test on some of my hosts. I have tried adding nosslcert to the host line after the testip keyword. I have tried to drop the test using the ./.xymon 120.0.0.1 "drop HOSTNAME sslcert" command Nothing is working. Ted This electronic message contains information generated by the USDA solely for the intended recipients. Any unauthorized interception of this message or the use or disclosure of the information it contains may violate the law and subject the violator to civil or criminal penalties. If you believe you have received this message in error, please notify the sender and delete the email immediately.

3 7

--nongreen-colors=red,purple not displayed
by Mark.Falzon＠fujixerox.com 03 Mar '14

03 Mar '14

Xymon version: 4.3.5 Hi all, We generate an only RED or PURPLE page through the xymongen --nongreen-colors=red,purple option. The page generates fine however when a client that has a purple test alongside of a yellow test xymongen seems to drop the client from the generated list. The client will still be part of "Current non-green systems" with its Purple and Yellow tests A colleague of mine thinks he has found where our problem comes from. This is based off the 4.3.16 code base. We run xymongen with the -nongreencolor=red,purple flags to have an Operations view that hides yellow alerts, but is meant to display all red and purple alerts. Our issue is for a sample Xymon client we might have a single client side test go into a purple state and also have one or more yellow statuses and this client will not be displayed in the "nongreen" page at all. In process.c (line 44) the logic uses a simple greater than comparison to determine the maximum "nongreencolor" of the host, so in our sample, this would be set to COL_YELLOW. In pagegen.c at line 1152, where the each host is being inspected to see if it is a candidate for inclusion in a page, the logic for setting "useit" uses the xymongen option of nongreencolors (which is a bit-shifted 1's complement) against a computed binary version of the host_t->nongreencolor. In our sample case, this means that the host_t->nongreencolor=COL_YELLOW and is currently excluded as we are not including Yellow's. One method of resolving this which seems to work is to make the host_t->nongreencolor to be a bit-shifted one's complement of the all the entry_t->color of the tests. --- process.c.orig 2014-02-28 07:14:30.000000000 +1100 +++ process.c 2014-02-28 07:16:17.000000000 +1100 @@ -41,10 +41,9 @@ if (e->propagate && (e->color > color)) color = e->color; oldage &= e->oldage; - if (e->propagate && (e->color > nongreencolor) && (strstr(nongreenignores, e->column->listname) == NULL)) { - nongreencolor = e->color; + if (e->propagate && (strstr(nongreenignores, e->column->listname) == NULL)) { + nongreencolor = ( (1 << e->color ) | nongreencolor ); } - if (e->propagate && e->alert && (e->color > criticalcolor)) { criticalcolor = e->color; } --- pagegen.c.orig 2014-01-20 08:49:42.000000000 +1100 +++ pagegen.c 2014-02-28 07:06:33.000000000 +1100 @@ -1152,7 +1152,7 @@ if (h->hostentry->nonongreen || (nongreennodialups && h->hostentry->dialup)) useit = 0; else - useit = (( (1 << h->hostentry->nongreencolor) & nongreencolors ) != 0); + useit = (( h->hostentry->nongreencolor & nongreencolors ) != 0); break; case PAGE_CRITICAL: @@ -1172,12 +1172,18 @@ break; } - if (useit) { + if (useit) { host_t *newhost, *walk; switch (summarytype) { case PAGE_NONGREEN: - if (h->hostentry->nongreencolor > nongreenpage.color) nongreenpage.color = h->hostentry->nongreencolor; + dbgprintf("Setting highest page colour, currently on %s\n",h->hostentry); + // Work out what the highest colour number is in the nongreencolor - I'm sure there is a more elegant way! + int tempcolour = 0; + if ( h->hostentry->nongreencolor > 31 ) { tempcolour=5; } + else if ( h->hostentry->nongreencolor > 15 ) { tempcolour=4; } + else if ( h->hostentry->nongreencolor > 7 ) { tempcolour=3; } + if ( tempcolour > nongreenpage.color) nongreenpage.color = tempcolour; break; case PAGE_CRITICAL: if (h->hostentry->criticalcolor > nongreenpage.color) nongreenpage.color = h->hostentry->criticalcolor; Thanks all Mark *********************************************************************************** This e-mail, including any attachments to it, may contain confidential and/or personal information. If you have received this e-mail in error, you must not copy, distribute, or disclose it, use or take any action based on the information contained within it. Please notify the sender immediately by return e-mail of the error and then delete the original e-mail. The information contained within this e-mail may be solely the opinion of the sender and may not necessarily reflect the position, beliefs or opinions of the organisation on any issue. This email has been swept for the presence of computer viruses known to the organisation's anti-virus systems. ***********************************************************************************

2 1

Swap two Xymon master sites to one master and one slave
by pto＠linuxbog.dk 03 Mar '14

03 Mar '14

Hi all I have made many Xymon/Hobbit server over the years - all as their own master. One thing I have not understood is how I can keep one machine as a slave and it should then send the CPU, memory etc data to the other If I have two machines each with their bb-hosts: 10.0.0.1: 10.0.01 host1 #bbd ssh 10.0.0.2 host2 # ssh 10.0.0.2: 10.0.02 host2 #bbd ssh 10.0.0.1 host1 # ssh How can I turn 10.0.0.2 into a slave, which sends cpu, disk, files etc to 10.0.0.1? And 10.0.0.1 should then display for both machines. Best -- Peter Toft <pto at linuxbog.dk>

2 1

xymon client pidfile
by feld＠feld.me 03 Mar '14

03 Mar '14

What is the historical reason for the xymon client pidfile to be named clientlaunch.`hostname`.pid ? Can that be overridden somehow? It makes it hard to do anything with the pidfile if the name is not static. We often override the hostname reported back to Xymon with --hostname=foo and parsing that out if --hostname exists is going to be a nightmare Thanks

2 1

Xymon home page suggestion
by solitaryr＠gmail.com 03 Mar '14

03 Mar '14

Henrik, If this is already there I'm not seeing it. I think it would be beneficial to have a link under the "Xymon Project" menu that points to the stable "Changes" folder so folks can see without downloading the latest source what changes have been committed. thoughts? =G=

4 3

Problem with "xymon download filename"
by abs＠shadymint.com 03 Mar '14

03 Mar '14

Hi, I have 4.3.13 and recently encountered 'too many open files' after a while using the download command. This came up before in relation to 4.3.5 and there was a patch posted here :- http://comments.gmane.org/gmane.comp.monitoring.hobbit/28865 but the patch is missing so I cannot tell if it made it into the subsequent releases. Can anyone confirm? Thanks -- Andy

2 1

Multiple schedules for DOWNTIME not displaying in info
by bb4＠richter-it.net 03 Mar '14

03 Mar '14

forwarding to the list as well. > ---------- Ursprüngliche Nachricht ---------- > Von: "bb4 at richter-it.net" <bb4 at richter-it.net> > An: Mike Burger <mburger at bubbanfriends.org> > Datum: 12. Februar 2014 um 11:11 > Betreff: Re: [Xymon] Multiple schedules for DOWNTIME not displaying in info > > > > Mike Burger <mburger at bubbanfriends.org> hat am 11. Februar 2014 um 19:59 > > geschrieben: > > > > > > > -----BEGIN PGP SIGNED MESSAGE----- > > > Hash: SHA1 > > > > > > Hi Mike, > > > > > > I had this problem too. > > > It seems, if you only have one scheduled downtime then spaces are no > > > problem but if you have more than one ... well it helps replacing the > > > spaces with underscores '_'. At least it worked for me and all scheduled > > > downtimes now show up on top of the info page. > > > > > > HTH > > > Torsten > > > > > > On 11.02.2014 17:41, Mike Burger wrote: > > >> I've got a server for which I want 2 different DOWNTIME schedules...one > > >> for "cpu" and one for "procs". > > >> > > >> The example on the hosts.cfg man page > > >> (http://www.xymon.com/xymon/help/manpages/man5/hosts.cfg.5.html) > > >> indicates > > >> that the proper syntax for DOWNTIME is: > > >> > > >> DOWNTIME=columns:day:starttime:endtime:cause[,columns:day:starttime:endtime:cause] > > >> > > >> Following that example, I've set it up like so: > > >> > > >> <ip address> <hostname> # OS:aix > > >> disk,files,cpu,memory,msgs,ports,procs > > >> NOCOLUMNS:inode DOWNTIME=procs:*:2159:0030:"Nightly dwmain and dwstage > > >> database backups",cpu:*:0000:2359:"CPU Utilization due to daily jobs" > > >> > > >> But when I go to the info page, I see: > > >> > > >> Planned downtime: All days:2159:0030 (status:procs) (cause:Nightly > > >> dwmain > > >> and dwstage database backups) > > >> > > >> and at the bottom: > > >> > > >> Other tags: OS:aix disk,files,cpu,memory,msgs,ports,procs > > >> NOCOLUMNS:inode > > >> ,cpu:*:0000:2359:CPU Utilization due to daily jobs > > >> > > >> Should I not see both parts of the DOWNTIME line in the "Planned > > >> downtime" > > >> area on the client's info page? I've noticed this behavior in both > > >> 4.3.15 > > >> and 4.3.16 (I'm only trying to add this multi-schedule for the first > > >> time, > > >> today). > > > > Thanks, Torsten. > > > > Did the schedules work, though, with the spaces in place (i.e. was it just > > an issue with the aesthetics) or did putting the the underscores ("_") in > > cause the schedule to properly function, too? > > > > -- > > Mike Burger > > http://www.bubbanfriends.org > > > > "It's always suicide-mission this, save-the-planet that. No one ever just > > stops by to say 'hi' anymore." --Colonel Jack O'Neill, SG1 > > > > > > Hi Mike, > > as far as I remember the schedules did not work. At least the second one > didn't. > So it was more than an aesthetics issue. > > Cheers > Torsten >

3 4