I installed the post-RC2 patch and 30 minutes later hobbitd went purple. The last entry in hobbitd.log is: 2005-02-17 15:43:56 hobbitd servername MACHINE='server,domain,com' not listed in bb-hosts, dropping hobbitd status
I checked bb-hosts, and it's in there: 1.2.3.4 server.domain.com # testip BBDISPLAY BBPAGER BBNET bbd ssh http://server.domain.com /hobbit/ https://1.2.3.4 smtp apache:http://127.0.0.1/server-status?auto
Did I botch the patch?
Thiesfeld, Gordon wrote:
I installed the post-RC2 patch and 30 minutes later hobbitd went purple. The last entry in hobbitd.log is: 2005-02-17 15:43:56 hobbitd servername MACHINE='server,domain,com' not listed in bb-hosts, dropping hobbitd status
I checked bb-hosts, and it's in there: 1.2.3.4 server.domain.com # testip BBDISPLAY BBPAGER BBNET bbd ssh http://server.domain.com /hobbit/ https://1.2.3.4 smtp *apache:http://127.0.0.1/server-status?auto * Did I botch the patch?
Shouldn't that be "apache=http://127.0.0.1/server-status?auto" ? If you have a typo maybe its making hobbitd crash...it seems particular about the apache keyword.
-Charles
On Thu, Feb 17, 2005 at 04:23:35PM -0600, Thiesfeld, Gordon wrote:
I installed the post-RC2 patch and 30 minutes later hobbitd went purple. The last entry in hobbitd.log is: 2005-02-17 15:43:56 hobbitd servername MACHINE='server,domain,com' not listed in bb-hosts, dropping hobbitd status
I checked bb-hosts, and it's in there: 1.2.3.4 server.domain.com # testip BBDISPLAY BBPAGER BBNET bbd ssh http://server.domain.com /hobbit/ https://1.2.3.4 smtp apache:http://127.0.0.1/server-status?auto
Someone else said he had the same problem - quite odd, but there must be something wrong somewhere. I'm also a bit puzzled why this has popped up with the RC2 release, because there were no changes that should change this behaviour.
Did I botch the patch?
Unlikely.
If you could send me your bb-hosts and hobbitserver.cfg files, I'll see if I can make it happen on my own system. That would ease debugging.
Regards, Henrik
On Thu, Feb 17, 2005 at 11:59:16PM +0100, Henrik Stoerner wrote:
On Thu, Feb 17, 2005 at 04:23:35PM -0600, Thiesfeld, Gordon wrote:
I installed the post-RC2 patch and 30 minutes later hobbitd went purple.
If you could send me your bb-hosts and hobbitserver.cfg files, I'll see if I can make it happen on my own system. That would ease debugging.
Thanks Gordon, with your config it happens on my system as well. And it's such a silly bug that I'm ashamed to admit I wrote it :-/
Henrik
We're looking at upgrading our current BB installation and hopbbit looks like a winner. Out current alerting is done via SNMP to an external site that generates trouble tickets and SMS messages from the SNMP trap data.
Are there any plans to have SNMP alerting as an option?
Regards Geoff Steer Firstwave technology P/L
On Sat, Feb 19, 2005 at 01:00:23AM +1100, Geoff Steer wrote:
We're looking at upgrading our current BB installation and hopbbit looks like a winner. Out current alerting is done via SNMP to an external site that generates trouble tickets and SMS messages from the SNMP trap data.
Are there any plans to have SNMP alerting as an option?
I don't have any immediate plans for this. It would probably not be terribly difficult to come up with a script to send out an SNMP trap, which Hobbit can run to do alerts - which is my main excuse for not adding SNMP trap support to Hobbit right now.
Sometime later this year, I'll probably look into adding some SNMP support to Hobbit, in the form of an SNMP module that can query SNMP-enabled devices for status information, thereby integrating support for SNMP devices into Hobbit. (SNMP has always been somewhat of a kludge in BB, you have to craft some ad-hoc script to call snmpget or snmpwalk, and parse the output in the script). When that happens, it will probably be pretty simple to implement a native SNMP alert mechanism in Hobbit.
But don't hold your breath ... right now, setting up an alert script to send out SNMP traps is the way to get something going.
Regards, Henrik
On Fri, 2005-02-18 at 15:45 +0100, Henrik Stoerner wrote:
On Sat, Feb 19, 2005 at 01:00:23AM +1100, Geoff Steer wrote:
Sometime later this year, I'll probably look into adding some SNMP support to Hobbit, in the form of an SNMP module that can query SNMP-enabled devices for status information, thereby integrating support for SNMP devices into Hobbit.
Hopefully snmp V3 support...
(SNMP has always been somewhat of a kludge in BB, you have to craft some ad-hoc script to call snmpget or snmpwalk, and parse the output in the script).
Right now I use bbmrtg.pl for this. mrtg is a fine polling engine, but it does have scaling issues. And it really needs to be database driven and self-reconfiguring.
When that happens, it will probably be pretty simple to implement a native SNMP alert mechanism in Hobbit.
But don't hold your breath ... right now, setting up an alert script to send out SNMP traps is the way to get something going.
I'm not holding my breath, but would like to eventually see snmp V3 support, which mrtg doesn't have currently, due to limitations of the perl module they use to poll.
-- Daniel J McDonald, CCIE # 2495, CNX Austin Energy
dan.mcdonald at austinenergy.com
On Fri, Feb 18, 2005 at 06:07:53PM -0600, Daniel J McDonald wrote:
On Fri, 2005-02-18 at 15:45 +0100, Henrik Stoerner wrote:
On Sat, Feb 19, 2005 at 01:00:23AM +1100, Geoff Steer wrote:
Sometime later this year, I'll probably look into adding some SNMP support to Hobbit, in the form of an SNMP module that can query SNMP-enabled devices for status information, thereby integrating support for SNMP devices into Hobbit.
Hopefully snmp V3 support...
Oh, I don't plan to implement a full SNMP client by myself. The Net-SNMP library is already there, and supports both SNMP v1, v2c and v3.
Henrik
participants (5)
-
dan.mcdonald＠austinenergy.com
-
gsteer＠firstwave.com.au
-
GThiesfeld＠webmd.net
-
henrik＠hswn.dk
-
jonescr＠cisco.com