As a workaround I did a a discovery.php -all every 4 hours. But the situation is just a little better.
So in summary here whats happens.
2017-12-28 06:01:47 Interface DELETED mark removed 2017-12-28 08:33:00 cron runs discovery.php -all
after discovery was complete - the situation didn't change. The interface is alive and the Observium shows traffic graphs. But it not appeared in alert checker.
Afterwhile checking a log i see this:
2017-12-28 10:37:02 Interface changed: [ifName] '' -> 'mgmt0'; [ifAlias] '' -> 'VPC Keep-alive link'; [ifHighSpeed] '' -> '1000'; [ifPromiscuousMode] '' -> 'false'; [ifConnectorPresent] '' -> 'true' 2017-12-28 10:31:22 Interface changed: [ifName] 'mgmt0' -> NULL; [ifAlias] 'VPC Keep-alive link' -> NULL; [ifHighSpeed] '1000' -> NULL; [ifPromiscuousMode] 'false' -> NULL; [ifConnectorPresent] 'true' -> NULL
still no info in alert checker.
But the strange thing is that in some alert checkers that interface appears. But if i do "regenerate" alert it comes to checker again and for sometime it's OK.
How could I improve the situation or give you more information - maybe this is somekind of bug ?
On Tue, Dec 19, 2017 at 8:52 AM, Edvinas K edvinas.email@gmail.com wrote:
- could you advice which files to put in the cron - to make auto
regenerate works ? Seems like that .php alone doesnt work
#auto regenrate 0 */7 * * * root /opt/observium/html/pages/alert_regenerate.inc.php
- i remember when someone said that after /opt/observium/discovery.php -h
all, there should be autoregeneration to - how to check if that really works ?
thank you
On Mon, Dec 18, 2017 at 6:06 PM, Adam Armstrong adama@memetic.org wrote:
You could run it half an hour after discovery ends or something.
No risk really, it just might be database intensive.
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=11327 On 18 Dec 2017, at 14:54, Edvinas K <edvinas.email <%3Ca>@gmail.com target=_blank>edvinas.email@gmail.com> wrote:
since only after the manual "alert regenrate" the missing ports starts to appear, i want to ask you.
Do you see any risks in putting the /opt/observium/html/pages/alert_regenerate.inc.php script in a cron tab and run in for example every ~3 hours ?
Thanks
On Fri, Dec 15, 2017 at 12:00 PM, Edvinas K edvinas.email@gmail.com wrote:
update:
after i DO:
/opt/observium/discovery.php -h jayXXX
##### Finished discovery run at 2017-12-15 09:57:03 #####
o Devices Discovered 1 o Discovery Time 249.8 secs o Memory usage 34MB (peak: 34MB) o MySQL Usage Cell[547/0.319s] Row[1739/1.934s] Rows[96/0.117s] Column[0/0s] Update[82/0.164s] Insert[2104/7.92s] Delete[6/0.023s]
i got this information which says that SNMP walk is OK.
And after that in port event log i see:
2017-12-15 10:55:30 Interface was marked as DELETED
why is that ?
On Wed, Dec 13, 2017 at 10:44 AM, Edvinas K edvinas.email@gmail.com wrote:
p.s after "alert regeneration" for sometime it's ok.
Somewhere i saw that after automatic disovery (discovery.php -h all) ? alerts should be regenerated ? is this still true?
Maybe there are some other command line to do only alarm regeneration
- for example every 3 hours.
On Wed, Dec 13, 2017 at 10:02 AM, Edvinas K edvinas.email@gmail.com wrote:
so where coud be a problem ?
Poller.php polls in quite normal time:
[2017/12/13 07:37:30 +0000] poller.php(28605): /opt/observium/poller.php: jaynetxxx - 1 devices polled in 148.7 secs [2017/12/13 07:42:22 +0000] poller.php(30304): /opt/observium/poller.php: jaynetxxx - 1 devices polled in 140.3 secs [2017/12/13 07:47:28 +0000] poller.php(2525): /opt/observium/poller.php: jaynetxxx - 1 devices polled in 147.1 secs [2017/12/13 07:52:27 +0000] poller.php(6010): /opt/observium/poller.php: jaynetxxx - 1 devices polled in 145.6 secs
Disovery process (which is snmpwalk ?) completes the job in: 235.2 seconds
Device SNMP seetings: Max repetitions is set to: 20 .
For example the the port is marked as deleted:
but the traffic graphs are still shown:
[image: Inline image 1]
When device is discovering (discovery -h) im seeing that the particular port is on:
| Protocol | ifName | Remote: hostname | port | platform | version | +----------+------------------ +----------------------------- ---------+-------------------- -----+----------------------+- ------------------------------ -----------+ | CDP | mgmt0 | JAYNET01A | mgmt0 | N5K-C5548UP | Cisco Nexus Operating System (NX-OS) ... |
So where could be the problem ?
On Tue, Dec 12, 2017 at 4:33 PM, Edvinas K edvinas.email@gmail.com wrote:
> /opt/observium/poller-wrapper. py --host jayxxx > /opt/observium/poller-wrapper. py:568: Warning: Data truncated for > column 'process_name' at row 1 > cursor.execute(p_query, (pid,ppid,processname,uid,comm > and,s_time)) > INFO: starting the poller at 2017/12/12 15:29:01 with 16 threads > /usr/bin/env php /opt/observium/poller.php -q -h 223 >> /dev/null > 2>&1 > INFO: starting alerter.php for 223 > INFO: finished alerter.php for 223 > INFO: worker Thread-1 finished device 223 in 141 seconds > INFO: poller-wrapper.py poller --host jayxxx processed 1 devices in > 141.60 seconds with 16 threads, load average (5min) 1.95 > > Number of rows updated: 1 > > here the observium.log information: > > > infra [root@observ001prvitx logs]# cat observium.log | grep jayxxx > | tail -30 > [2017/12/12 12:27:38 +0000] poller.php(26069): > /opt/observium/poller.php: jayxxx - 1 devices polled in 156.5 secs > [2017/12/12 12:32:38 +0000] poller.php(27576): > /opt/observium/poller.php: jayxxx - 1 devices polled in 155.8 secs > [2017/12/12 12:37:39 +0000] poller.php(24853): > /opt/observium/poller.php: jayxxx - 1 devices polled in 157.5 secs > [2017/12/12 12:42:36 +0000] poller.php(31743): > /opt/observium/poller.php: jayxxx - 1 devices polled in 155.1 secs > [2017/12/12 12:47:35 +0000] poller.php(918): > /opt/observium/poller.php: jayxxx - 1 devices polled in 154.1 secs > [2017/12/12 12:52:34 +0000] poller.php(2685): > /opt/observium/poller.php: jayxxx - 1 devices polled in 152.7 secs > [2017/12/12 12:57:33 +0000] poller.php(6267): > /opt/observium/poller.php: jayxxx - 1 devices polled in 151.7 secs > [2017/12/12 13:02:30 +0000] poller.php(11093): > /opt/observium/poller.php: jayxxx - 1 devices polled in 148.2 secs > [2017/12/12 13:07:37 +0000] poller.php(14342): > /opt/observium/poller.php: jayxxx - 1 devices polled in 155.0 secs > [2017/12/12 13:12:20 +0000] poller.php(17116): > /opt/observium/poller.php: jayxxx - 1 devices polled in 138.3 secs > [2017/12/12 13:17:12 +0000] poller.php(18797): > /opt/observium/poller.php: jayxxx - 1 devices polled in 129.9 secs > [2017/12/12 13:22:29 +0000] discovery.php(18423): > /opt/observium/discovery.php: jayxxx - 1 devices discovered in 178.5 secs > [2017/12/12 13:22:47 +0000] poller.php(25318): > /opt/observium/poller.php: jayxxx - 1 devices polled in 166.0 secs > [2017/12/12 13:27:35 +0000] poller.php(27070): > /opt/observium/poller.php: jayxxx - 1 devices polled in 153.2 secs > [2017/12/12 13:32:22 +0000] poller.php(28740): > /opt/observium/poller.php: jayxxx - 1 devices polled in 140.1 secs > [2017/12/12 13:37:19 +0000] poller.php(31747): > /opt/observium/poller.php: jayxxx - 1 devices polled in 137.1 secs > [2017/12/12 13:42:18 +0000] discovery.php(14545): > /opt/observium/discovery.php: jayxxx - 1 devices discovered in 206.4 secs > [2017/12/12 13:42:39 +0000] poller.php(2225): > /opt/observium/poller.php: jayxxx - 1 devices polled in 157.0 secs > [2017/12/12 13:47:31 +0000] poller.php(5992): > /opt/observium/poller.php: jayxxx - 1 devices polled in 149.2 secs > [2017/12/12 13:52:21 +0000] poller.php(8767): > /opt/observium/poller.php: jayxxx - 1 devices polled in 139.3 secs > [2017/12/12 13:55:21 +0000] poller.php(18678): > /opt/observium/poller.php: jayxxx - 1 devices polled in 129.5 secs > [2017/12/12 13:57:16 +0000] poller.php(10836): > /opt/observium/poller.php: jayxxx - 1 devices polled in 134.1 secs > [2017/12/12 14:02:34 +0000] poller.php(14434): > /opt/observium/poller.php: jayxxx - 1 devices polled in 152.4 secs > [2017/12/12 14:07:16 +0000] poller.php(16199): > /opt/observium/poller.php: jayxxx - 1 devices polled in 134.4 secs > [2017/12/12 14:12:22 +0000] poller.php(18143): > /opt/observium/poller.php: jayxxx - 1 devices polled in 140.1 secs > [2017/12/12 14:17:17 +0000] poller.php(19794): > /opt/observium/poller.php: jayxxx - 1 devices polled in 134.8 secs > [2017/12/12 14:22:20 +0000] poller.php(22296): > /opt/observium/poller.php: jayxxx - 1 devices polled in 138.0 secs > [2017/12/12 14:27:13 +0000] poller.php(25090): > /opt/observium/poller.php: jayxxx - 1 devices polled in 131.4 secs > [2017/12/12 14:31:22 +0000] poller.php(9089): > /opt/observium/poller.php: jayxxx - 1 devices polled in 141.0 secs > [2017/12/12 14:32:30 +0000] poller.php(26759): > /opt/observium/poller.php: jayxxx - 1 devices polled in 148.5 secs > infra [root@observ001prvitx logs]# > > > On Tue, Dec 12, 2017 at 4:26 PM, Markus Klock < > markus@best-practice.se> wrote: > >> and no eventlog entry about snmpwalk timeout? >> >> /Markus >> >> 2017-12-12 15:21 GMT+01:00 Edvinas K edvinas.email@gmail.com: >> >>> I did a manual poller-wrapper and sicovery on it - but still, i >>> dont see a port (entity) in alert checker. >>> >>> Alo attaching a Screenshot: >>> >>> >>> >>> [image: Inline image 1] >>> >>> On Tue, Dec 12, 2017 at 3:58 PM, Markus Klock < >>> markus@best-practice.se> wrote: >>> >>>> Can you screenshot the Performance Data of that device? Cog-wheel >>>> at the top right on the device-page. >>>> Cisco Nexus have a pretty bad SNMP-stack. >>>> /Markus >>>> >>>> 2017-12-12 14:45 GMT+01:00 Edvinas K edvinas.email@gmail.com: >>>> >>>>> I did a manual discover (Discovery time - 206.3) but it still >>>>> marked as deleted: >>>>> >>>>> 2017-12-12 14:42:05 Interface was marked as DELETED < after my >>>>> discover. >>>>> 2017-12-12 14:26:32 Interface DELETED mark removed >>>>> 2017-12-12 14:22:16 Interface was marked as DELETED >>>>> 2017-12-12 14:06:33 Interface DELETED mark removed >>>>> 2017-12-12 14:01:56 Interface was marked as DELETED >>>>> 2017-12-12 09:56:36 Interface changed: [ifAlias] '' -> 'vPC >>>>> Keep-alive link' >>>>> >>>>> What's wrong ? >>>>> >>>>> >>>>> >>>>> On Tue, Dec 12, 2017 at 3:22 PM, Adam Armstrong < >>>>> adama@observium.org> wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> Alert (and group) associations are rebuilt directly after a >>>>>> device discovery, so if during a discovery some ports are removed or their >>>>>> ifType/ifAlias is returned black because of timed out snmp, they'll drop >>>>>> off this checker. >>>>>> >>>>>> Do you see the ifType/Alias being changed in the entity's logs? >>>>>> >>>>>> adam. >>>>>> >>>>>> On 2017-12-12 13:09:15, Edvinas K edvinas.email@gmail.com >>>>>> wrote: >>>>>> Hello, >>>>>> >>>>>> I've created alarm checker: >>>>>> >>>>>> os equals nxos >>>>>> hostname match *NET* >>>>>> >>>>>> ifType equals ethernetCsmacd >>>>>> ifAlias match *vPC Keep-alive link* >>>>>> >>>>>> >>>>>> And after a while i noticed that some entities where >>>>>> disapeared. The device was reachable and etc. >>>>>> I did a "rebuild' and they're reappeared again. What could be >>>>>> the cause ? NXOS ? :) >>>>>> ______________________________ _________________ observium >>>>>> mailing list observium@observium.org >>>>>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >>>>>> >>>>>> >>>>>> ______________________________ _________________ >>>>>> observium mailing list >>>>>> observium@observium.org >>>>>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >>>>>> >>>>>> >>>>> >>>>> ______________________________ _________________ >>>>> observium mailing list >>>>> observium@observium.org >>>>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >>>>> >>>>> >>>> >>>> ______________________________ _________________ >>>> observium mailing list >>>> observium@observium.org >>>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >>>> >>>> >>> >>> ______________________________ _________________ >>> observium mailing list >>> observium@observium.org >>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >>> >>> >> >> ______________________________ _________________ >> observium mailing list >> observium@observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >> >> >
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium