NetXMS Support Forum

Hello!

You should take binary package for IPSO. Latest available version is 0.2.23 at https://www.netxms.org/download/archive/agent_packages/nxagent-0.2.23-ipso.apkg. If you run it as


nxagent-0.2.23-ipso.apkg prefix=/opt/netxms


it should install agent under /opt/netxms. I'll also built 0.2.31 version for IPSO.

Best regards,
Victor

Po kakim-to prichinam bilo otkrito mnogo sessij na agent, kotorie ne zakrivalis' po time-out'u. Eto mozet bit' bug v agente. Esche takoe mozet sluchitsja, esli agent ispol'zuetsja kak proxy libo dlja proverki otkritih portov subagentom portcheck.nsm. Esli ispol'zuetsja portcheck ili proxy, to luchshe uvelichit' kolichestvo sessij - znachenie po umolchaniju 32, mozno poprobovat' podnjat' do 100. Esli portcheck ili proxy ne ispol'zujutsja, to pohoze na bug (ili kto-to dejstvitel'no postojanno podkljuchaetsja k agentu, setevoj skaner naprimer). Esli situacija povtoritsja, prishlite pozalujsta rezul'tat komandi netstat -an i konfig agenta.

Dlja NetWare toze est' agent, i on podderzivaet parametri File.Size i File.Count. V novoj versii pojavitsja takze parametr File.Count s viborkoj po razmeru i vremeni modifikacii.

Dobrij den'!

Znachenie peremennoj counter mozno vstavit' v stroku vot tak:

CUSTOM_MESSAGE="Interface Ethernet0/" . counter . " changed state to DOWN";

A sam skript mozno sokratit' do

sub main()
{
   CUSTOM_MESSAGE = "Interface Ethernet0/" . int32(($2 - 386) / 128) . " changed state to DOWN";
   return 1;
}

ili, esli trebuetsja proverka chto $2 - eto 386 + chislo, kratnoe 128, i ne previshaet 3458, to:

sub main()
{
   if (($2 >= 3458) || (($2 - 386) % 128 != 0))
      return 0;
   CUSTOM_MESSAGE = "Interface Ethernet0/" . int32(($2 - 386) / 128) . " changed state to DOWN";
   return 1;
}

Yes, version of the agent can be different. So far, any server version in 0.2.x branch can work with any agent version, including very old 0.1.x version agents.

Best regards,
Victor

Hello!

1. It depends on how customized they should be. In current version of management console you can place various information windows (like alarm browser, history graphs, etc.) inside main console window and save layout. You can create as many saved layouts as you wish. We also working on a completely new console application, which is based on Eclipse platform, which will allow you to quickly develop custom plugins.

2. Yes - there are alarm view in console.

3. Both ways are possible - depending on what changes / extensions you need. System was designed to be modular and easily extendable. You can create your own modules for server and agent (subagents). There are client libraries in C and Java - they can be used for integration with other systems or to create your own client applications. However, there are one major problem - there are almost no documentation on API, so you will have to ask questions on forum and read source code (or you can order custom development from us).

Best regards,
Victor

Разобрался с этой проблемой, в ближайшие дни выложу версию 0.2.31, будет инсталлироваться. Если надо срочно, пишите, могу вечером собрать и выложить отдельно инсталлер для x64 агента.

Разобрался с этой проблемой, в ближайшие дни выложу версию 0.2.31, будет инсталлироваться. Если надо срочно, пишите, могу вечером собрать и выложить отдельно инсталлер для x64 агента.

Try to remove all files from <install_prefix>/share/netxms/packages

Best regards,
Victor

A v kakom meste ispol'zuetsja skript? V event processing policy? Esli da, to on dolzen vozvraschat' 0 ili 1 - 0 oznachaet chto sobitie pod pravilo ne podhodit - 1 - chto podhodit. Custom message nado zadavat' cherez peremennuju CUSTOM_MESSAGE, bez znaka dollara, i potom mozete ispol'zovat' eto znachenie v alarmah i action'ah cherez macros %M. Naprimer:

sub main()
{
   if ($1==770)
      CUSTOM_MESSAGE="FastEthernet 0/3";
   return 1;
}

I ja podumaju kak sdelat', chtobi ne pisat' if dlja kazdogo interfejsa.

That means that for nodes with id 84 and 86 collected data will not be stored. I suggest to recreate these tables by the following queries:

CREATE TABLE idata_84 (item_id integer not null,idata_timestamp integer not null,idata_value varchar(255) not null);
CREATE INDEX idx_idata_84_item_id ON idata_84(item_id);
CREATE INDEX idx_idata_84_id_timestamp ON idata_84(item_id,idata_timestamp);

CREATE TABLE idata_86 (item_id integer not null,idata_timestamp integer not null,idata_value varchar(255) not null);
CREATE INDEX idx_idata_86_item_id ON idata_86(item_id);
CREATE INDEX idx_idata_86_id_timestamp ON idata_86(item_id,idata_timestamp);

Best regards,
Victor

Also, there are a bug registered in MySQL bugtracker - http://bugs.mysql.com/bug.php?id=15667 which seems similar to your problem. One of possible solutions mentioned in comments is to increase value of server variable innodb_buffer_pool_size (may be useful if you use InnoDB database, not MyISAM).

Also, if manual DELETE FROM statements will not work, try to clear tables with

TRUNCATE TABLE syslog;
TRUNCATE TABLE event_log;
TRUNCATE TABLE snmp_trap_log;
TRUNCATE TABLE audit_log;

Best regards,
Victor

Hi!

Try to connect to your database with mysql client and clear log tables manually by executing the following queries:

DELETE FROM syslog;
DELETE FROM event_log;
DELETE FROM snmp_trap_log;
DELETE FROM audit_log;

Then run upgrade process again.

Best regards,
Victor

I have fixed it. Agent now will correctly handle rename of monitored file (which usually happens during log rotation). I plan to release version 0.2.31 in a few days, it will contain this fix.

Best regards,
Victor

Спасибо за trace, решил проблему. В течении нескольких дней выложу версию 0.2.31, которая будет корректно работать с трапами версии 3.