MSP Various problems
Hi There.
We constantly have an issue with one of our routers, as you can see by the attached screen shot the routers internal IP address in the example is 192.168.254.254 but OpManager has incorrectly changed its IP to 208.69.32.130 all on its own.
If I manually click on edit and put the correct IP in again, it is ok for a while, but then if it doesn't respond a few times sometimes it will just change IP back to the 208. address, which is a opendns IP.
Any idea why this is happening? and how do we stop it from randomly changing our devices IP address in it's database?
We have OpManager MSP edition, and we also constantly have issues with the central server whereby the graphs stop showing any images on the servers snapshot pages, at the same time the devices next poll time shows inaccurately as hours behind the current time.
The only way to start it working again is to stop the two opmanager services and then because it doesn't always stop we have to kill the mysqld-nt.exe process manually and then the java.exe iexplore.exe and wrapper.exe processes if they are still running, after stopping them all we run the repairdb.bat file then start services again, after that the images start working again.
Just like I have seen a few other people post, when we need to do the above, when the repairdb.bat file is ran, our database regularly has issues needing to be fixed, as in "clients is using or hasn't closed table properly".
It happens so regularly now that we've taken the "pause" line out of the repairdb.bat file and created a batch file to do all of those steps above which is scheduled to run every night, but we still occasionally have to run it again during the day.
Another thing that occasionally happens with remote probes is that they will die if they cannot contact the central server for x amount of time, so we have to dial in to the customers networks and restart and fix their probes just the same process as we need to do on the central, so again - on some sites we've also scheduled a script that does a similar stop and repair process to stop us having to do these fixes manually.
On top of all that, we find that even with the windows account being used being a domain admin, which we set the same on all sites, we also set the same SNMP and WMI settings on all servers, sometimes adding the exchange app monitor won't monitor any information or detect what it's supposed to, or it won't detect the database locations etc, other times it works fine, these are all on windows 2003 servers with the same service pack so is not an inconsistancy between O/S's.
But as well as being able to report both the edb and stm sizes together, it would be handy to display and change the exchange database path after the app monitor has been added, as customers do sometimes need to move the edb file paths, it's a shame there is no way to even remove the exchange application manager itself without removing the whole server and adding it all in again.
Just one final thing, we also have on our own network our exchange server being monitored, but it is reporting some inaccurate stats, if you see the attached screen shot of our exchange info page, it shows the sending message queue as 3010 emails, this is not correct and in our exchange system manager there are no stuck emails.
These are just a few annoying quirks and makes the software seem buggy the longer we use it, as the central server it is on was loaded fresh and has no other software or jobs and since being loaded has ran nothing else apart from OpManager central for monitoring our customers, but yet it still has these little quirks and inconsistencies.
We are currently up to 17 probes now, the plan was to keep purchasing and rolling out probes until all our customers critical servers were being monitored, but we are not rolling out any more until we can get the current glitches ironed out or workarounds sorted, as it's causing us more work than we would ideally like.
Any ideas on any of the points mentioned would be much appreciated.
Regards
Simon Murawski
We constantly have an issue with one of our routers, as you can see by the attached screen shot the routers internal IP address in the example is 192.168.254.254 but OpManager has incorrectly changed its IP to 208.69.32.130 all on its own.
If I manually click on edit and put the correct IP in again, it is ok for a while, but then if it doesn't respond a few times sometimes it will just change IP back to the 208. address, which is a opendns IP.
Any idea why this is happening? and how do we stop it from randomly changing our devices IP address in it's database?
We have OpManager MSP edition, and we also constantly have issues with the central server whereby the graphs stop showing any images on the servers snapshot pages, at the same time the devices next poll time shows inaccurately as hours behind the current time.
The only way to start it working again is to stop the two opmanager services and then because it doesn't always stop we have to kill the mysqld-nt.exe process manually and then the java.exe iexplore.exe and wrapper.exe processes if they are still running, after stopping them all we run the repairdb.bat file then start services again, after that the images start working again.
Just like I have seen a few other people post, when we need to do the above, when the repairdb.bat file is ran, our database regularly has issues needing to be fixed, as in "clients is using or hasn't closed table properly".
It happens so regularly now that we've taken the "pause" line out of the repairdb.bat file and created a batch file to do all of those steps above which is scheduled to run every night, but we still occasionally have to run it again during the day.
Another thing that occasionally happens with remote probes is that they will die if they cannot contact the central server for x amount of time, so we have to dial in to the customers networks and restart and fix their probes just the same process as we need to do on the central, so again - on some sites we've also scheduled a script that does a similar stop and repair process to stop us having to do these fixes manually.
On top of all that, we find that even with the windows account being used being a domain admin, which we set the same on all sites, we also set the same SNMP and WMI settings on all servers, sometimes adding the exchange app monitor won't monitor any information or detect what it's supposed to, or it won't detect the database locations etc, other times it works fine, these are all on windows 2003 servers with the same service pack so is not an inconsistancy between O/S's.
But as well as being able to report both the edb and stm sizes together, it would be handy to display and change the exchange database path after the app monitor has been added, as customers do sometimes need to move the edb file paths, it's a shame there is no way to even remove the exchange application manager itself without removing the whole server and adding it all in again.
Just one final thing, we also have on our own network our exchange server being monitored, but it is reporting some inaccurate stats, if you see the attached screen shot of our exchange info page, it shows the sending message queue as 3010 emails, this is not correct and in our exchange system manager there are no stuck emails.
These are just a few annoying quirks and makes the software seem buggy the longer we use it, as the central server it is on was loaded fresh and has no other software or jobs and since being loaded has ran nothing else apart from OpManager central for monitoring our customers, but yet it still has these little quirks and inconsistencies.
We are currently up to 17 probes now, the plan was to keep purchasing and rolling out probes until all our customers critical servers were being monitored, but we are not rolling out any more until we can get the current glitches ironed out or workarounds sorted, as it's causing us more work than we would ideally like.
Any ideas on any of the points mentioned would be much appreciated.
Regards
Simon Murawski