|
Hallo,
dass untenstehend beschriebene ist ein wirklich spezielles
Problem. Dafür gibt es eigentlich keine Lösung, ausser man
baut den Teststring so auf, dass auf jedem Server eine
Datenbankverbindung erzwungen wird.
Gruß, Jacob
> -----Original Message-----
> From: Alois Treindl [mailto:alois@xxxxxxxx]
> Sent: Sunday, June 10, 2001 10:41 PM
> To: 'lvs-users@xxxxxxxxxxxxxxxxxxxxxx'
> Subject: monitoring tools - ldirectord
>
>
> What are you people using to automatically monitor an LVS cluster
> (the real servers, not the director?)
>
> I have just installed ldirectord from the ultramonkey project and
> it works as described, but I have a problem to solve:
>
> I have 4 realservers which run http services, named w1, w2, w3, w4.
>
> ldirectord manages them nicely, i.e. removes/adds via ipvsadm if
> a server is turned off/on.
>
> If all 4 are down, it switches to the fallback server (on the director
> itself).
>
> I have a special requirement, however: server w1 is the dbms server,
> and if this one is down, it must be treated as if all realservers are
> down, because the other realservers cannot function properly
> without w1
>
> ldirectord.cf does not seem to have such a overriding feature
> which gives
> one realserver special importance.
>
> Should I modify ldirectord (which is written in Perl and not
> very large)
> or is there another convenient solution for my problem?
>
> Alois
>
>
> _______________________________________________
> LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
> Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
> or go to http://www.in-addr.de/mailman/listinfo/lvs-users
>
|
