NetXMS Support Forum

A v logah servera nichego net pro oshibki pri starte (naprimer chto ne udalos' sdelat' bind na port 514)? I esche - mozno li prislat' neskol'ko obrazcov soobschenij - esli server ih prinimaet to edinstvennaja prichina po kotoroj oni mogut ne pojavljatsja v loge eto oshibka pri parsinge.

Sudja po logu, configure otrabotal normal'no. A gde vse lomaetsja? Esche odin obschij sovet, inogda pomogaet na komercheskih UNIX'ah - poprobovat' sobrat' s --disable-iconv.

Znachit netxms_ip ne propisan v parametre Servers. Nado ego tam propisat' (zelatel'no kak MasterServers).

Esli ExternalParameter opredelen kak SipOrig(*), to v Parameter nado pisat'

SipOrig(192.168.100.22)

Ostal'noe vrode vse pravil'no. Dolzen rabotat.
Mozno eshe zapustit' agenta s kljuchom -D i posmotret' na kakoj parametr prihodjat zaprosi.

Vilozil promezutochnij reliz 0.2.20-dev2 na https://www.netxms.org/download/rc/ . Nado delat' upgrade i serveru, i consoli. VNIMANIE: esli posle upgrade'a redaktirovat' DCI konsol'ju ot versii 0.2.19, to nastroiki "proxy node" budut sbrosheni v znachenie po umolchaniju.

Uze pochti pojavilas' Ja sdelal novij attribut u DCI - "proxy node" - t.e. DCI konfiguritsja na odnoj node a znachenie real'no sprashivaetsja u nodi vistavlennoj kak proxy. Protestiruju i vilozu promezutochnij reliz - mozno budet probovat'.

Quote from: drew on October 15, 2007, 12:37:50 PM

Рассылка по нескольким e-mail адресам, разделённым ";" или пробелом. Можно, конечно, сделать несколько экшенов, как я сейчас и делаю, но была бы фича полезна.

Sdelal, v CVS versii uze est'. Adresa nado razdeljat' ";". Toze samoe rabotaet i v rassilke SMS-ov. Esli ochen' hochetsja etu funkcional'nost' v tekushej versii, to mozno zamenit' fail src/server/core/actions.cpp na tot kotorij v attache i peresobrat' server.

T.e. nuzno chtobi script ispolnjalsja na hoste A no znachenie shlo v DCI dlja hosta B? Esli tak, to normal'nogo reshenija poka net. Edinstvenno chto mogu predlozit' - eto ispol'zovat' push DCI i ne vizivat' script s servera, a zapuskat' ego po cron'u i ispol'zovat' nxpush dlja peredachi novogo znachenija DCI na server monitoringa.

I'll try to setup Microsoft SQL Server 2005 Express Edition to chek it myself. Please wait till Monday's evening.

Best regards,
Victor

Standartnimi sredstvami NetXMS poka nel'zja. Odnako mozno delat' selecti prjamo iz bazi na usnovje polja platform_name i/ili snmp_oid v tablice nodes.

V osnovnom budet rashodovatsja pamjat', tak chto rekomenduju minimum 1GB RAM. Krome togo, objazatel'no nado budet uvelichit' po sravneniju s defaultnimi kolichestvo pollerov (cherez konfiguracionnie parametri servera). Esli baza dannih na etoj-ze mashine, to esche dobavit' pamjati dlja db engine. Po processoru osobih trebovanij net, ljuboj sovremennij proc spravitsja. Nu i setevoj traffic mozet bit' dostatochno bol'shim, vozmozno ponadobitsja uvelichenie intervalov mezdu oprosami sostojanija i sborom dannih.

Hello!

Do you have TCP/IP enabled on SQL server? NetXMS can use only TCP/IP connection.

Best regards,
Victor

Based on config table content:

Looks like a problem with active discovery. Difference between Windows and Linux is that on Windows if you try to ping address 0.0.0.0 you will get an error, but on Linux it will return normal reply from loopback interface. So probably you have address 0.0.0.0 in your active discovery configuration, and on Linux server constantly find additional nodes. Could you also send me content of address_lists table?

Best regards,
Victor

Icons attached. From tomorrow, they also should be in CVS snapshot: https://www.netxms.org/download/netxms-cvs.tar.gz
lld.dsp has no use in it's current condition, so you can just delete it from workspace.

Best regards,
Victor

Teoreticheski mozet, poskol'ku vse oni peredajutsja na klientskuju mashinu pri sinhronizacii. Ja poprobuju sdelat' testovuju VM s bol'shim kolichestvom logicheskih interfeisov i posmotrju chto poluchaetsja.