NetXMS Support Forum

Network map format has been changed in 1.1.x branch and no longer supported in legacy Windows console. You should use Java console (nxmc) for work with 1.1.x server.

Best regards,
Victor

Hi!

This caused by a known bug/limitation in MySQL. Until we create a workaround, you can upgrade database manually. If you have a backup of version 218, then you can:

1. Manually execute query


CREATE TABLE images (guid varchar(36) not null,   mimetype varchar(64) not null, name varchar(255) not null, category varchar(255) not null,   protected integer default 0,      PRIMARY KEY(guid));


2. Run nxdbmgr upgrade -X

To fix already upgraded database, manually execute the following queries:


CREATE TABLE images (guid varchar(36) not null,   mimetype varchar(64) not null, name varchar(255) not null, category varchar(255) not null,   protected integer default 0,      PRIMARY KEY(guid));

INSERT INTO images (guid, name, category, mimetype, protected) VALUES
  ('092e4b35-4e7c-42df-b9b7-d5805bfac64e', 'Service', 'Network Objects', 'image/png', 1);
INSERT INTO images (guid, name, category, mimetype, protected) VALUES
  ('1ddb76a3-a05f-4a42-acda-22021768feaf', 'ATM', 'Network Objects', 'image/png', 1);
INSERT INTO images (guid, name, category, mimetype, protected) VALUES
  ('7cd999e9-fbe0-45c3-a695-f84523b3a50c', 'Unknown', 'Network Objects', 'image/png', 1);
INSERT INTO images (guid, name, category, mimetype, protected) VALUES
  ('904e7291-ee3f-41b7-8132-2bd29288ecc8', 'Node', 'Network Objects', 'image/png', 1);
INSERT INTO images (guid, name, category, mimetype, protected) VALUES
  ('b314cf44-b2aa-478e-b23a-73bc5bb9a624', 'HSM', 'Network Objects', 'image/png', 1);
INSERT INTO images (guid, name, category, mimetype, protected) VALUES
  ('ba6ab507-f62d-4b8f-824c-ca9d46f22375', 'Server', 'Network Objects', 'image/png', 1);
INSERT INTO images (guid, name, category, mimetype, protected) VALUES
  ('bacde727-b183-4e6c-8dca-ab024c88b999', 'Router', 'Network Objects', 'image/png', 1);
INSERT INTO images (guid, name, category, mimetype, protected) VALUES
  ('f5214d16-1ab1-4577-bb21-063cfd45d7af', 'Printer', 'Network Objects', 'image/png', 1);
INSERT INTO images (guid, name, category, mimetype, protected) VALUES
  ('f9105c54-8dcf-483a-b387-b4587dfd3cba', 'Switch', 'Network Objects', 'image/png', 1);


Best regards,
Victor

Hi!

I put compiled binaries of NetXMS server with SQLite database driver here: https://www.netxms.org/download/dev/netxms-1.1.7-aix.tar.gz. It is intended to be unpacked into /opt directory. It may require additional libraries, please let me know if you got errors when try to run it.

Best regards,
Victor

Hi!

Sorry, I don't fully understand the question. Do you mean that you need to get some information from some Java agent via SNMP?

Best regards,
Victor

Hi!

Yes, the main problem with advanced schedule is that you cannot specify year in schedule. Otherwise within a year you can create very flexible schedules - don't forget that you can specify more than one schedule for DCI. For example, monitoring from 15.01 9:00 till 10.03 14:59 can be defined as follows:

* 9-23 15 1 *
* * 16-31 1 *
* * * 2 *
* * 1-9 3 *
* 0-14 10 3 *

1 - covers period 15.01 9:00 - 23:59
2 - covers period 16.01 00:00 - 31.01 23:59
3 - covers entire February
4 - covers period 01.03 00:00 - 09.03 23:59
5 - covers period 10.03 00:00 - 10.03 14:59

And I don't think that using transformation script here is a misuse - it is what they was intended for.

Best regards,
Victor

You can get them from subversion repository: https://svn.netxms.org/public/netxms/trunk/src/java. Release of 1.1.8 was planned to the end of December, but we are a bit late on schedule. I hope that it will be ready for release this week.

Best regards,
Victor

Hi!

If you are not interested in checking that new content matches some exact text, but just to check for fact of change, then my proposed solution with diff thresholds will work. However, if you have to check that new content is available and matching some specific text/regexp, you have to create new DCI for each upcoming content change, and use advanced scheduling to execute it at specific time. With advanced scheduling, you will be able to execute the check at exact time or in given short interval. You can also use Java or C API to automate DCI creation if you have some external application which knows what the content will be.

Best regards,
Victor

Hi!

The problem is that agent interprets last colon as a separator between script name and polling interval. It is not a problem on UNIX, but became a problem on Windows. Try to specify in config as:

ExternalParametersProvider = D:\topsec\hp_eva_collect_data.pl:60

(60 is a polling interval in seconds, you can change it to suit your needs). Then it should work.

Best regards,
Victor

Hi!

Quote from: danieln on January 01, 2012, 05:12:18 PM
1. My main goal is to be able to push events from a java application into the NMS. As my java application emits JMX notifications, I want to catch them and using your java-api send a status event to netxms. Do you consider that a valid solution? Is there an online example of java api use for statuses and events?

Yes, this is a valid solution. There are no online examples, but you can use Junit tests in Java API source tree as examples of how to do simple operations with Java API. And feel free to ask any questions regarding API on the forum. As for sending events, appropriate API call will be available in 1.1.8.

Quote from: danieln on January 01, 2012, 05:12:18 PM
2. Another requirement is to be able to execute commands from the network map on hosts and components. Is it possible to define such commands to be invoked from a context menu on the map? (e.g. open a web browser with a specific URL)

Yes, it's possible. You can define various "object tools", including for open specific URL or executing command either on management server, administrator's machine, or remote node with NetXMS agent. Those tools accessible anywhere in console where you see objects - on network maps, in object browser, in alarm list, etc.

Best regards,
Victor

Hi!

What database you plan to use?

Best regards,
Victor

Hi!

What is exact purpose of monitoring this web site? If you need to detect content changes on any time except allowed, you can just use "diff" threshold, and in event processing policy skip event generated by threshold if it occurs within valid content change time interval.

Best regards,
Victor

Hi!

If you run script from command line, you should get output similar to this:

root@netxms:~# /opt/netxms/bin/hp_eva_collect_data.pl
EVA.Disk[7].loops.loop[2].loopname=loopb
EVA.DiskShelf[1].diskslot[2].name=diskslot2
EVA.Disk[7].objecttype=disk
EVA.DiskShelf[1].diskslot[6].diskid=080720072004000CCA65B9400000000000000000
EVA.Controller[1].datablocksize=512
EVA.Disk[3].rssindex=6
EVA.Controller[1].controllername=Controller 1
EVA.Disk[1].disktype=fibre_channel_disk
EVA.DiskShelf[1].diskslot[11].name=diskslot11
EVA.DiskShelf[1].cooling.fan[2].firmware=80.1
EVA.DiskShelf[1].diskslot[5].diskwwn=2004-000c-ca2a-f588-0000-0000-0000-0000
EVA.Disk[8].operationalstate=attention
EVA.Disk[7].diskgroupname=Default Disk Group
EVA.Disk[3].operationalstatedetail=member_attention

... and so on. Do you get it?

Best regards,
Victor

Hi!

Alarm generation and actions are different things. For generating alarm you do not need an action - it's just special attributes in event processing policy rule. To set rule to generate alarm, do the following (assuming you already have rule object called "rule" of type EventProcessingPolicyRule):


rule.setFlags(rule.getFlags() | EventProcessingPolicyRule.GENERATE_ALARM);
rule.setAlarmMessage("%m");  // it's an example, you can specify your real alarm message here
rule.setAlarmSeverity(0);   // specify your severity
rule.setAlarmKey("");   // specify alarm key


and to set rule to terminate alarm:


rule.setFlags(rule.getFlags() | EventProcessingPolicyRule.GENERATE_ALARM);
rule.setAlarmSeverity(Severity.UNMANAGED);   // Severity UNMANAGED indicates alarm termination rule
rule.setAlarmKey("");   // specify alarm key


Also, you can take a look at property pages in nxmc-epp plugin in management console.

Best regards,
Victor

No, it is not necessary. Empty main() will be created by script compiler if it was not defined explicitly.

Best regards,
Victor

Just added "rename script" function to the Java console.

Quote from: lindeamon on December 25, 2011, 01:37:19 PM
when will the java console include all the old features of the native console ?

I hope that version 1.1.8 will be able to replace legacy console completely. There are no much functions left.

Best regards,
Victor