![](https://secure.gravatar.com/avatar/948a18f2bb45baf8cafea289dac502e4.jpg?s=120&d=mm&r=g)
Hi,
I recently installed Observium CE and noticed some very weird behavior.
Observium monitors currently 3 devices: 1 Debian Server + 1 Debian workstation + 1 Ubuntu workstation. The Debian workstation is shutdown overnight and rebooted in the morning, nicely shown in the Event logging:
Shutdown: *<date time> System Device status changed to Down (ping)* Reboot: *<date time> System Device status changed to Up <date time> Reboot Device rebooted after <hours min sec> * That's all fine. But now the weird thing is that for the Server the Event log says almost at the same time:
*<date time> System Device status changed to Down (snmp)*
and when the workstation is rebooted, the Server Event log says at almost the same time:
*<date time> System Device status changed to Up*
Graphs for the server don't get updated/have data for the time the Debian workstation is down. The Ubuntu workstation runs 24/7 and graphs get properly updated during time the Debian workstation is down.
I would expect that what happened with the workstation should have no effect on the data collection of the Server.
Details:
Observium CE 0.14.4.5229 OS Debian 7 (Server + Workstation) Ubuntu 10.40 LTS (Workstation 24/7)
How do I get this working properly?
Thanx,
Pieter
![](https://secure.gravatar.com/avatar/948a18f2bb45baf8cafea289dac502e4.jpg?s=120&d=mm&r=g)
All,
I worked out what is causing the problem: NFS shares!
The server has _NFS shares_ which are located on the workstation which is shutdown. As soon as the workstation is down,**the 1st *snmpwalk* will not complete and timeout somewhere halfway. Any subsequent *snmpwalk* commands will give a timeout straightaway. This explains why there is no data recorded until the workstation is up again. At that point, the_NFS shares _are restored and *snmpwalk* will successfully run.
To fix this problem I have added the following line in my */etc/snmp/snmpd.conf*:
*skipNFSInHostResources true*
After restarting the SNMPD service NFS shares are exclude and observium is recording the server details while the workstation is down.
Cheers,
Pieter
On 26/05/2014 18:43, overige@vanderpoel.com.au wrote:
Hi,
I recently installed Observium CE and noticed some very weird behavior.
Observium monitors currently 3 devices: 1 Debian Server + 1 Debian workstation + 1 Ubuntu workstation. The Debian workstation is shutdown overnight and rebooted in the morning, nicely shown in the Event logging:
Shutdown: *<date time> System Device status changed to Down (ping)* Reboot: *<date time> System Device status changed to Up <date time> Reboot Device rebooted after
<hours min sec> * That's all fine. But now the weird thing is that for the Server the Event log says almost at the same time:
*<date time> System Device status changed to Down (snmp)*
and when the workstation is rebooted, the Server Event log says at almost the same time:
*<date time> System Device status changed to Up*
Graphs for the server don't get updated/have data for the time the Debian workstation is down. The Ubuntu workstation runs 24/7 and graphs get properly updated during time the Debian workstation is down.
I would expect that what happened with the workstation should have no effect on the data collection of the Server.
Details:
Observium CE 0.14.4.5229 OS Debian 7 (Server + Workstation) Ubuntu 10.40 LTS (Workstation 24/7)
How do I get this working properly?
Thanx,
Pieter
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
![](https://secure.gravatar.com/avatar/21caf0a08d095be7196a1648d20942be.jpg?s=120&d=mm&r=g)
Pieter,
Ooh! I like that toggle, as I have the same issue when an NFS host is down... snmp blocks on the disk free stats for NFS. Thanks for the pointer!
Tom
On 29/05/2014 03:38, overige@vanderpoel.com.au wrote:
All,
I worked out what is causing the problem: NFS shares!
The server has _NFS shares_ which are located on the workstation which is shutdown. As soon as the workstation is down,**the 1st *snmpwalk* will not complete and timeout somewhere halfway. Any subsequent *snmpwalk* commands will give a timeout straightaway. This explains why there is no data recorded until the workstation is up again. At that point, the_NFS shares _are restored and *snmpwalk* will successfully run.
To fix this problem I have added the following line in my */etc/snmp/snmpd.conf*:
*skipNFSInHostResources true*
After restarting the SNMPD service NFS shares are exclude and observium is recording the server details while the workstation is down.
Cheers,
Pieter
On 26/05/2014 18:43, overige@vanderpoel.com.au wrote:
Hi,
I recently installed Observium CE and noticed some very weird behavior.
Observium monitors currently 3 devices: 1 Debian Server + 1 Debian workstation + 1 Ubuntu workstation. The Debian workstation is shutdown overnight and rebooted in the morning, nicely shown in the Event logging:
Shutdown: *<date time> System Device status changed to Down (ping)* Reboot: *<date time> System Device status changed to Up <date time> Reboot Device rebooted after
<hours min sec> * That's all fine. But now the weird thing is that for the Server the Event log says almost at the same time:
*<date time> System Device status changed to Down (snmp)*
and when the workstation is rebooted, the Server Event log says at almost the same time:
*<date time> System Device status changed to Up*
Graphs for the server don't get updated/have data for the time the Debian workstation is down. The Ubuntu workstation runs 24/7 and graphs get properly updated during time the Debian workstation is down.
I would expect that what happened with the workstation should have no effect on the data collection of the Server.
Details:
Observium CE 0.14.4.5229 OS Debian 7 (Server + Workstation) Ubuntu 10.40 LTS (Workstation 24/7)
How do I get this working properly?
Thanx,
Pieter
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
![](https://secure.gravatar.com/avatar/48bfe696ac1cbf068a4de2b752e281c6.jpg?s=120&d=mm&r=g)
Iirc there is bug net snmp about Nfs shares, fixed in recent version
On 29 мая 2014 г., at 5:38, "overige@vanderpoel.com.au" overige@vanderpoel.com.au wrote:
All,
I worked out what is causing the problem: NFS shares!
The server has NFS shares which are located on the workstation which is shutdown. As soon as the workstation is down, the 1st snmpwalk will not complete and timeout somewhere halfway. Any subsequent snmpwalk commands will give a timeout straightaway. This explains why there is no data recorded until the workstation is up again. At that point, the NFS shares are restored and snmpwalk will successfully run.
To fix this problem I have added the following line in my /etc/snmp/snmpd.conf:
skipNFSInHostResources true
After restarting the SNMPD service NFS shares are exclude and observium is recording the server details while the workstation is down.
Cheers,
Pieter
On 26/05/2014 18:43, overige@vanderpoel.com.au wrote: Hi,
I recently installed Observium CE and noticed some very weird behavior.
Observium monitors currently 3 devices: 1 Debian Server + 1 Debian workstation + 1 Ubuntu workstation. The Debian workstation is shutdown overnight and rebooted in the morning, nicely shown in the Event logging:
Shutdown: <date time> System Device status changed to Down (ping) Reboot: <date time> System Device status changed to Up <date time> Reboot Device rebooted after <hours min sec>
That's all fine. But now the weird thing is that for the Server the Event log says almost at the same time:
<date time> System Device status changed to Down (snmp)
and when the workstation is rebooted, the Server Event log says at almost the same time:
<date time> System Device status changed to Up
Graphs for the server don't get updated/have data for the time the Debian workstation is down. The Ubuntu workstation runs 24/7 and graphs get properly updated during time the Debian workstation is down.
I would expect that what happened with the workstation should have no effect on the data collection of the Server.
Details:
Observium CE 0.14.4.5229 OS Debian 7 (Server + Workstation) Ubuntu 10.40 LTS (Workstation 24/7)
How do I get this working properly?
Thanx,
Pieter
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
participantes (3)
-
Nikolay Shopik
-
overige@vanderpoel.com.au
-
Tom Laermans