Hi,
I'm running Xymon since 6 years (4.3.17 atm) on Debian 7.8 3.2.0-4-amd64 Since 1 month now, every night, between 0h30 or 2h am at +/- 30 min, around 30 hosts become unreachable :
Fri Jan 29 01:16:38 2016 conn NOT ok : DNS lookup failed Unable to resolve hostname foo.bar.local System unreachable for 3 poll periods (170 seconds) green 0.0.0.0 is alive (0.02 ms) [<- 127.0.0.1]
- Got around 500 monitored hosts and looks like the same hosts are lost every single night.
- Those monitored hosts are not necessary on the same network, not the same OS.
- We cross monitored the same hosts and the other monitoring tool doesn't have report the DNS outage.
- I ran a DNS lookup every seconds on the Hobbit server several days and it never reported a DNS outage.
- I don't have any crontab installed on the server who could disturb Xymon.
- Nothing strange in the Xymon logs nor the server logs, no memory leaks or CPU overloaded.
- The rest of the day, Xymon server behavior is normal.
- What I've done on the server 1 month ago ? I don't know, no system upgrade or so.
- I had DNSMASQ acting like a cache, I disabled it : same issue
- /etc/resolv.conf is quite light : search bar.local, next line : nameserver IP.OF.OUR.DNS.SERVER1, just like other servers
The issue could be anywhere : inside or outside the server, Xymon or not... I have to confess, I'm running out of ideas to find the issue, is anyone here may have some leads, I will be thankful !
Have a nice day!
Hi L-M-J,
can you exclude that this behavior is coming from any network device like a switch or default gateway?
Regards Christian
Christian Becker IT-Services
Christian.Becker at rhein-zeitung.net<mailto:Christian.Becker at rhein-zeitung.net>
Mittelrhein-Verlag GmbH August-Horch-Straße 28 D-56070 Koblenz Verleger und Geschäftsführer: Walterpeter Twer Reg.-Gericht Koblenz HRB 121 Finanzamt Koblenz Str.Nr. 22 65 10 285 2 www.rhein-zeitung.de<http://www.rhein-zeitung.de/>
Von: Xymon [mailto:xymon-bounces at xymon.com] Im Auftrag von L-M-J Gesendet: Freitag, 29. Januar 2016 08:57 An: Xymon at xymon.com Betreff: [Xymon] Xymon disruption every night!
Hi,
I'm running Xymon since 6 years (4.3.17 atm) on Debian 7.8 3.2.0-4-amd64 Since 1 month now, every night, between 0h30 or 2h am at +/- 30 min, around 30 hosts become unreachable :
Fri Jan 29 01:16:38 2016 conn NOT ok : DNS lookup failed Unable to resolve hostname foo.bar.local System unreachable for 3 poll periods (170 seconds) green 0.0.0.0 is alive (0.02 ms) [<- 127.0.0.1<http://127.0.0.1>]
- Got around 500 monitored hosts and looks like the same hosts are lost every single night.
- Those monitored hosts are not necessary on the same network, not the same OS.
- We cross monitored the same hosts and the other monitoring tool doesn't have report the DNS outage.
- I ran a DNS lookup every seconds on the Hobbit server several days and it never reported a DNS outage.
- I don't have any crontab installed on the server who could disturb Xymon.
- Nothing strange in the Xymon logs nor the server logs, no memory leaks or CPU overloaded.
- The rest of the day, Xymon server behavior is normal.
- What I've done on the server 1 month ago ? I don't know, no system upgrade or so.
- I had DNSMASQ acting like a cache, I disabled it : same issue
- /etc/resolv.conf is quite light : search bar.local, next line : nameserver IP.OF.OUR.DNS.SERVER1, just like other servers
The issue could be anywhere : inside or outside the server, Xymon or not... I have to confess, I'm running out of ideas to find the issue, is anyone here may have some leads, I will be thankful !
Have a nice day!
Problems appears on VMs and physical servers and Lan and DMZ equipments. I don't see a link between those devices :-(
Le 29 janvier 2016 09:23:14 GMT+01:00, Becker Christian <christian.becker at rhein-zeitung.net> a écrit :
Hi L-M-J,
can you exclude that this behavior is coming from any network device like a switch or default gateway?
Regards Christian
Christian Becker IT-Services
Christian.Becker at rhein-zeitung.net<mailto:Christian.Becker at rhein-zeitung.net>
Mittelrhein-Verlag GmbH August-Horch-Straße 28 D-56070 Koblenz Verleger und Geschäftsführer: Walterpeter Twer Reg.-Gericht Koblenz HRB 121 Finanzamt Koblenz Str.Nr. 22 65 10 285 2 www.rhein-zeitung.de<http://www.rhein-zeitung.de/>
Von: Xymon [mailto:xymon-bounces at xymon.com] Im Auftrag von L-M-J Gesendet: Freitag, 29. Januar 2016 08:57 An: Xymon at xymon.com Betreff: [Xymon] Xymon disruption every night!
Hi,
I'm running Xymon since 6 years (4.3.17 atm) on Debian 7.8 3.2.0-4-amd64 Since 1 month now, every night, between 0h30 or 2h am at +/- 30 min, around 30 hosts become unreachable :
Fri Jan 29 01:16:38 2016 conn NOT ok : DNS lookup failed Unable to resolve hostname foo.bar.local System unreachable for 3 poll periods (170 seconds) green 0.0.0.0 is alive (0.02 ms) [<- 127.0.0.1<http://127.0.0.1>]
- Got around 500 monitored hosts and looks like the same hosts are lost every single night.
- Those monitored hosts are not necessary on the same network, not the same OS.
- We cross monitored the same hosts and the other monitoring tool doesn't have report the DNS outage.
- I ran a DNS lookup every seconds on the Hobbit server several days and it never reported a DNS outage.
- I don't have any crontab installed on the server who could disturb Xymon.
- Nothing strange in the Xymon logs nor the server logs, no memory leaks or CPU overloaded.
- The rest of the day, Xymon server behavior is normal.
- What I've done on the server 1 month ago ? I don't know, no system upgrade or so.
- I had DNSMASQ acting like a cache, I disabled it : same issue
- /etc/resolv.conf is quite light : search bar.local, next line : nameserver IP.OF.OUR.DNS.SERVER1, just like other servers
The issue could be anywhere : inside or outside the server, Xymon or not... I have to confess, I'm running out of ideas to find the issue, is anyone here may have some leads, I will be thankful !
Have a nice day!
-- Envoyé de mon appareil Android avec K-9 Mail. Veuillez excuser ma brièveté.
My intention was the figure out if the network connection of the Xymon server itself has a problem… For example, if your Xymon server is hardware, then it has a wired network interface that is connected to a network switch. That’s your link between the Xymon server and all of your other VMs and physical servers.
participants (2)
-
christian.becker＠rhein-zeitung.net
-
linuxmasterjedi＠free.fr