- Xymon - Xymon mailman web

proc test fails
by chrisp＠lightstoneauto.co.za 09 May '19

09 May '19

Good day On one of my very busy severs (Centos7, tomcat8.5) the proc test fails I noticed that on the SERVICES=procs page for the server only about 4 line are shown from the ps output. The problem only happens when the server is busy. I added more ram to the server but the result is the same. Memory usage on server is ok (Reported by xymon): Used Total Percentage Real/Physical 18464M 24070M 70% Actual 9041M 7999M 37% Swap/Page 39M 7999M 0% The xymon client was installed from the Terabithia repo xymon-client-4.3.28-1.el7.x86_64 What do I need to change in my configs to remedy the problem please. Kind regards [https://lsc-assets.scdn1.secure.raxcdn.com/lssig/AUTOlogoW.png] Chris Pretorius [https://lsc-assets.scdn1.secure.raxcdn.com/lssig/phone.png] 0872367740 [https://lsc-assets.scdn1.secure.raxcdn.com/lssig/email.png] chrisp at lightstone.co.za<mailto:chrisp at lightstone.co.za> [https://lsc-assets.scdn1.secure.raxcdn.com/lssig/web.png] www.signio.co.za<http://www.signio.co.za> [https://lsc-assets.scdn1.secure.raxcdn.com/lssig/address.png] Centurion Square, 3rd Floor, cnr Heuwel and Gordon Hood rd, Centurion This communication is subject to Lightstone's email disclaimer<http://www.lightstone.co.za/portal/content/email-disclaimer.aspx>

3 4

how to add a specific http timeout
by JoseManuel.Blasco＠itemsp.es 07 May '19

07 May '19

Good Morning. I run Xymon 4.3.25 and I need to enable an alert for http 8280 when timeout is over 1 second. Can I do easily way in hosts or I have to create a new configuration as you explain in olders version? "No, this will be possible in the next release - but not in 4.3.x. What you can do is to use the "NET:foo" definition on this particular host in hosts.cfg to have it tested by a special instance of the xymonnet task: 1) Add the "--test-untagged" option to your current xymonnet command in tasks.cfg 2) Add "NET:slowtests" to this particular host in hosts.cfg 3) In a script, set XYMONNETWORK="slowtests" (remember to export it!) and then run the xymonnet command to do the test of this one host, with the different --timeout option 4) Add a new section to tasks.cfg to run the test script." Thanks. Best Regards. [JoseManuelBlasco]

2 1

RPMs - logrotate - duplicate entry
by john.horne＠plymouth.ac.uk 07 May '19

07 May '19

Hi, I'm just throwing this one out there as I haven't been able to think about it too much. With the Terabithia RPMs, on our Xymon server I have installed both the 'xymon' and 'xymon-client' packages. The problem is that they both install an '/etc/logrotate.d' file - called 'xymon' and 'xymon-client' respectively. Both files are exactly the same. However, when logrotate runs it actually throws up an error because it is trying to rotate the same file twice (because there are two logrotate files which reference it). This error is emailed to me: ==== Subject: Anacron job 'cron.daily' on <server name> /etc/cron.daily/logrotate: error: xymon-client:4 duplicate log entry for /var/log/xymon/acknowledge.log ==== My thoughts about this have only got as far as thinking that perhaps when the packages are installed, a pre or post installation check for the existence of both logrotate files would cause one of them to be deleted. (Since both files are exactly the same there is no need to install them both.) The downside would be if, perhaps, we later removed the 'xymon' package (maybe moved it to another server), we might then be left with no logrotate file on the server. The 'acknowledge.log' file exists but is empty. Not sure if that is relevant. John. -- John Horne | Senior Operations Analyst | Technology and Information Services University of Plymouth | Drake Circus | Plymouth | Devon | PL4 8AA | UK ________________________________ [http://www.plymouth.ac.uk/images/email_footer.gif]<http://www.plymouth.ac.uk/worldclass> This email and any files with it are confidential and intended solely for the use of the recipient to whom it is addressed. If you are not the intended recipient then copying, distribution or other use of the information contained is strictly prohibited and you should not rely on it. If you have received this email in error please let the sender know immediately and delete it from your system(s). Internet emails are not necessarily secure. While we take every care, University of Plymouth accepts no responsibility for viruses and it is your responsibility to scan emails and their attachments. University of Plymouth does not accept responsibility for any changes made after it was sent. Nothing in this email or its attachments constitutes an order for goods or services unless accompanied by an official order form.

2 1

Integer overflow in client-side files check
by abe＠deuxchevaux.org 03 May '19

03 May '19

Hi, occassionally determining which file to check takes so long that the file has been modified already after the "files" check started but before the file timestamp is actually checked. Hence the file might have a negative age. But since the variable where this is stored seems to be unsigned, this causes false positives due to an integer overflow: > red /nf/2019/05/03/201905032050 > File was modified 4294967291 seconds ago - should be <360 > File was modified 4294967291 seconds ago - should be <960 This probably only happens if the file to check first has to be found using backticks or $(...) -- which may take some time. In our case the according line in client-local.cfg looks like this: file:`ls -1 $(ls -1d /nf/2???/??/??/ | tail -1)* | tail -1` Timing examples: > $ time ls -1 $(ls -1d /nf/2???/??/??/ | tail -1)* | tail -1 > /nf/2019/05/03/201905032055 > > real 0m0.015s > user 0m0.003s > sys 0m0.004s I see several potential solutions in decreasing preference: A) Don't compare to the check startup time but always to the time at which the file timestamp is checked. B) Use a signed integer instead of an unsigned integer. C) Ignore timestamps newer than "now" when comparing with "<". Actually IMHO the proper fix would be to implement both, A and B. P.S.: Happens with 4.3.28 installed from Debian Stretch's xymon package. Haven't checked with the 4.3.29 branch, but there was nothing related listed in 4.3.29's Changes file so far. Kind regards, Axel -- PGP: 2FF9CD59612616B5 /~\ Plain Text Ribbon Campaign, http://arc.pasp.de/ Mail: abe at deuxchevaux.org \ / Say No to HTML in E-Mail and Usenet Mail+Jabber: abe at noone.org X https://axel.beckert.ch/ / \ I love long mails: https://email.is-not-s.ms/

1 0

PAGE/VPAGE
by jonas.hultstrom＠scania.com 03 May '19

03 May '19

Hi, Maybe a stupid thing (again)… In my hosts.cfg I have: page SAP_P P-systems 1.2.3.4 l4311p #NAME:PE1 SAP 5.6.7.8 l4317p #NAME:PS1 SAP When I use PAGE the "NAME" turns up on the html page: [cid:image001.png at 01D501BE.E6D8F590] but when I change to VPAGE the hostname is displayed. vpage SAP_P P-systems 1.2.3.4 l4311p #NAME:PE1 SAP 5.6.7.8 l4317p #NAME:PS1 SAP [cid:image002.png at 01D501BE.E6D8F590] Do you know if it's possible to use VPAGE with this feature? Jonas

1 0

Patch 4: extra graph options
by stef.coene＠docum.org 03 May '19

03 May '19

Hi, This patch adds 2 extra options when generating graphs. The option filter can be used to filter out the graphs so it contains only certain lines. The second option is a filelist option that can be used to specify the rrd files that should be displayed in the graph. This is very handy when you want to integrate the graphs in an external reporting tool and want to control exactly the rrds used to create a graph. Only display /admin, / and /tmp file system: showgraph.sh?host=snsrv243&service=disk&filelist=,admin|,root|,tmp&action=menu Only display file systems with db in the mountpoint: showgraph.sh?host=snsrv243&service=disk&filter=db&action=menu REMARK: The filter option is not working 100%. The comments in the graph are all blank. I don't use this option a lot so I never bothered to figure out what the problem is. Stef

1 0

Patch 3: EVENTSTART variable when alerting
by stef.coene＠docum.org 03 May '19

03 May '19

Hi, This is a rather simple one and sets the variable EVENTSTART when sending out an alert. We use this with external scripting to match the exact event. The combination host + service + eventstart is an unique key. Stef

1 0

Patch 2: AIX Physical CPU and Entitled Capacity
by stef.coene＠docum.org 03 May '19

03 May '19

Hi, On AIX there are 2 numbers that are very useful in the vmstat output. They represent that Physical CPU used and the Entitled CPU capacity that the server uses. This patch allows these 2 numbers to be parsed and stored in the vmstat rrd. REMARK: if there is already an rrd, this rrd will miss these 2 extra counters. You can either remove the old rrd or update the rrd and add the 2 extra data sources. There are some perl scripts that can do this for you. An alternative is to create a new rrd file to hold these 2 numbers. Stef

1 0

Patch 1: AIX Actual memory
by stef.coene＠docum.org 03 May '19

03 May '19

Hi, AIX reports the Actual memory (AVM) and this is a very useful number. It's the amount of memory that's required by AIX and the running programs. The first patch filters the AVM out of the message and stores them in memactused. It also sets memacttotal. The second patch is needed because memactpct can be > 100 on AIX. Stef

1 0

Goodbye and thanks for all the fish
by EDSchminke＠Hormel.com 03 May '19

03 May '19

+1 We had been using "Big Brother" for years. We had also switched to a new fangled, whiz-bang monitoring solution and it just-- wasn't-- working out. We finally decided to pull the plug and switch back to Xymon because it just friggin' works. Good luck, I guess... we'll be here when you need us again. :) Erik D. Schminke | Associate Systems Programmer Hormel Foods Corporation | One Hormel Place | Austin, MN 55912 Phone: (507) 434-6817 edschminke at hormel.com | www.hormelfoods.com

5 4