NetXMS Support Forum

Hi,

getScriptLibrary() call intended to retrieve list of available scripts (because script source could be quite big). Source code for each script should be retrieved individually using getScript() method. The following script will show sources for all scripts:


scripts = session.getScriptLibrary()
for script in scripts:
    script = session.getScript(script.getId())
    print('id={0} name={1} source={2}'.format(
             script.getId()
            ,script.getName()
            ,script.getSource()
        )
    )


Best regards,
Victor

Hi,

can you share exact errors you have? What database driver you are using?

Best regards,
Victor

Can you please reproduce timeout issue again, with server running at debug level 7 and agent at debug level 6, and send me synchronized log files for period during synchronization?

Best regards,
Victor

It looks that from time to time database cannot cope with incoming data rate and so writer queue grows. You may need to fine tune your database server. But it seems to happen on few occasions so it should not be a root cause for data synchronization timeouts.

Best regards,
Victor

Hi,

please check object ID for MAINCR_GSX (you can find it on overview tab), then run the following command on NetXMS server command line:

nxadm -c "show topology <id>"

(replace <id> with actual node ID) and post it's output.

Best regards,
Victor

Hi,

what is your idata writer queue size? You can check it in server debug console with command "show queue". If you have historical graph for idata queue please also share it.
Also, how many DCIs you have? You can check it in server debug console with command "show stats".

Best regards,
Victor

Hi,

looks like server is locked on sending object updates to one of user sessions. We encounter same issue in another installation recently and already changed object update notification mechanism in current development branch, so there are good chance that upgrade to 2.2.9 will fix your issue.

Best regards,
Victor

Hi,

load on thread pool MAIN is definitely not normal. Could you please capture thread stack traces using attached script (you will need gdb installed)?

Best regards,
Victor

I was able to reproduce this issue. It is fixed in development branch and fix will be included into next release.

Best regards,
Victor

Hi,

could you please provide all steps to reproduce server crash? Or provide crash dump of the server?

Best regards,
Victor

Hi,

I don't think that currently it is possible without direct DB write. I can make an easy change to PushDCIData function to accept timestamp (or create separate function) to make such transformations possible.

Best regards,
Victor

Looks like you have MIB called AT-SMI-MIB missing. Check that you have MIB file with .txt extension starting with

AT-SMI-MIB DEFINITIONS ::= BEGIN

Best regards,
Victor

Добавим - скорее всего не в виде встроенной карты, а показа маркера в навигационном приложении по умолчанию. Сейчас есть функция "navigate to" для нод с выставленноми координатами.

Hi,

it's not possible yet.

Best regards,
Victor

Hi,

it's problem with reading data from database, not with storing. It is caused by bug in MariaDB client library https://jira.mariadb.org/browse/CONC-281. It is supposed to be fixed in latest client version, but could be not. Our MariaDB driver has check for client version that activates workaround - maybe it erroneously assumes version number where this issue was fixed. You can try to find the following lines in src/db/dbdrv/mariadb/mariadb.cpp:


/* workaround for MariaDB C Connector bug CONC-281 */
if (mysql_get_client_version() < 100209)  // for any client before 10.2.9
{
    mysql_stmt_store_result(hStmt->statement);
}


and comment out version check so it will look like this:

/* workaround for MariaDB C Connector bug CONC-281 */
//if (mysql_get_client_version() < 100209)  // for any client before 10.2.9
{
    mysql_stmt_store_result(hStmt->statement);
}


then recompile MariaDB driver.

Best regards,
Victor