NetXMS Forum

Greetings,

I am attempting to upgrade to 2.1 on CentOS 6.3, but I'm receiving this message during the configure process:

configure: error: libssh is required for SSH support. Check that it is installed or use --disable-ssh option

I am assuming that SSH would be a good thing to have operational, yes?  But I already have libssh installed:

yum install libssh2                                  Loaded plugins: fastestmirror
Setting up Install Process
Loading mirror speeds from cached hostfile
 * base: mirror.steadfast.net
 * extras: mirror.cs.vt.edu
 * updates: mirror.jax.hugeserver.com
Package libssh2-1.4.2-2.el6_7.1.x86_64 already installed and latest version
Nothing to do

Thanks in advance for your help!

-Kevin C.

OK, good news:

This command worked, with more parameters:

du -h --max-depth=1 | sort -n

I see now the problem is in /var/lib/mysql/

110516876 netxxx.millerxxx.com.err

This seems to be an error file, that is now massive?

-Kevin C.

Hi,

I ran du -h, and the only thing it came back with was "116K".

-Kevin C.

Greetings,

I am out of disk space on the NetXMS CentOS volume; and it is not the MySQL logs.  Does anyone have ideas on what to check first?

-Kevin C.

I did a restore on the CentOS server, so I'm now back up and running again on 1.2.17.  However, I would still like to upgrade.  Is it best to upgrade in incremental steps, instead of going directly to 2.0.8?  And if the nxdbmgr check is clear with no errors, shouldn't the database upgrade succeed?

I'm not sure how to contact Raden directly.

-Kevin C.

So, I attempted the upgrade to v2.0.8.  nxdbmgr check went fine, so I proceeded, although the db upgrade failed with 2 tables that needed a repair.  I ran the repair on the 2 tables (one was ALARMS, the other was EVENT_LOG), now I'm getting this:

[root@netmgmt netxms-2.0.8]# nxdbmgr upgrade
NetXMS Database Manager Version 2.0.8 Build 8078 (2.0.8-RELEASE) (NON-UNICODE)

Upgrading database...
Upgrading from version 387 to 388
SQL query failed (Duplicate column name 'dci_id'):
ALTER TABLE alarms ADD dci_id integer
Rolling back last stage due to upgrade errors...
Database upgrade failed
[root@netmgmt netxms-2.0.8]#

This is a production system, and I need to get it back online.  Thanks in advance for your quick response.

-Kevin C.

Greetings,

The NetXMS service won't start for some reason on CentOS.  I ran nxdbmgr check, and had to repair 2 MySQL tables, and also it fixed 17k + errors.  Now nxdbmgr check is all good, no errors, but the NetXMS service will not start.  The agent service does start.

Yes I know, I need to upgrade - but haven't had the time to do it yet.

-Kevin C.

Hi,

OK great, thank you!  Can I install the newer 2.1-M1 Windows Java management console, and for now run the stable 2.0.6 version on the CentOS server?  Or do console and server versions have to match?

-Kevin C.

Hi,

Can I upgrade directly from 1.2.17 (running on CentOS) to 2.0.6?  I'm thinking I should get this done, before upgrading to the 2.1 branch?

-Kevin C.

Hi,

Are we able to upgrade from Version 1.2.17 (running on CentOS) directly to version 2.0.3, using the standard upgrade process?

-Kevin C.

Thanks for the quick reply.  I checked MySQL first, and although binary logging was not enabled, the mysqld.log was huge!  I removed it, and all is good now.

I'll setup a log rotation script, and this problem should not re-occur.

-Kevin C.

Greetings,

We are running v 1.2.17 on Linux/MySQL, and it is out of disk space, again.  This is happening every 2-3 months.  What can be done to purge (to take care of the immediate problem), and also limit the size of logs, events, maybe auto remove objects that are no longer online, etc.?  I'm getting really good at expanding Linux partitions, but it's getting old.

Thanks in advance for the help!

-Kevin C.

Hi Victor,

I think you are correct, there were some /24 subnet objects.  Some of the nodes in those /24 subnets are bound to containers, and some are not.  But any node that is not bound to a container should be re-discovered if I remove the entire subnet, correct?  Assuming of course that my discovery filters are allowing those subnets to be searched.

-Kevin C.

OK, good news, this is now working in my test environment    A poll count of 5 rendered a node down alarm (and email) in ~8-9 minutes.  I thought it would be closer to 5 minutes, so not sure why it was longer, but I can adjust it down to get to 5 minutes if needed.

In working on this issue, the only thing I noticed is that on many nodes, I had a lot of interfaces named "lost_interface_####".  Not just on the nodes I was testing, but others as well.  I went thru the console using a view filter of "lost_in", and deleted all of those interfaces.

After I did that, I put the poll count of 5 into my test node interface, and it worked !  So maybe I had some database corruption?

-Kevin C.

The nodes responded to pings from the server.  So I took another look at my discovery settings, and then recalled I had filtered out the upper end of the IP address ranges of those networks when I initially installed NetXMS !

I modified those filters, saved it, and within 5 minutes, NetXMS discovered the missing nodes... chalk this up to pilot error

-Kevin C.

Hi,

It's quite easy, just add the devices as source objects to the built-in Event Processing Policy named "Show alarm when node is down".  Also, you will probably want to add them to the related policy named "Acknowledge node down alarm when node is up".

-Kevin C.

Hi,

No, that is not my intention for the final configuration.  I just want 5 minutes.

But I had changed it to 5 at first, and that made no difference - it went to "node down" status in ~60 seconds.  So I decided to put in 300 (just a random big number) to see if that would make any difference, and it did not.

-Kevin C.

Hi Victor,

I have testing by setting the node interface "required poll count" parameter to 300.  There is now a question mark on the node interface icon on the NetXMS console. Then, I disable the switch port that node is plugged into.  But no change, I still get a node down alert in approximately 60 seconds.

Is there anything else I need to check?

-Kevin C.

Greetings,

We are getting many of these alerts with version 1.2.17:

Invalid network mask 255.255.255.192 on interface "vlan1", should be 255.255.255.0

The /26 (255.255.255.192) is in fact the correct mask for this device; this particular node is a firewall that has VLAN's configured on a trusted port.

-Kevin C.

Greetings,

NetXMS 1.2.17 is not discovering all nodes, in 3 or 4 of our remote networks.  It discovered some nodes in those networks, but not all.  Is there a way to force a "re-discovery" of just a certain network?

-Kevin C.