Couchdb crashes, chef-server does not survive couchdb restart


#1

Hi all

I have experienced multi crashes of couchdb on my chef-server. I dont know why, there is no log, no trace no indication from couchdb.
My database is around 30 GB and the server is redhat 5.5 having 3 GB physical memory.

I heard that erlang sometime kill processes dueto lack of resource or other contention, and restart them automatically. But in my case
the couchdb process (mochiweb I think) that faciliates database access via http simply dies and I have restart couchdb entirely.

But the problem from Chef side is that Chef server can not re-connect to couchdb again and I have to restart chef-server-api then chef-server-webui

Anyone is facing similar problem?

Huy Le
Tecnología ING DIRECT
Administrador de sistema
Tel: +34 91 206 76 85

Por favor, no imprima este correo si no es necesario.


Attention:
The information contained in this message and or attachments is intended
only for the person or entity to which it is addressed and may contain
confidential and/or privileged material. Any review, retransmission,
dissemination or other use of, or taking of any action in reliance upon,
this information by persons or entities other than the intended recipient
is prohibited. If you received this in error, please contact the sender and
delete the material from any system and destroy any copies.


#2

On 24 February 2011 08:43, le.huy@ingdirect.es wrote:

My database is around 30 GB and the server is redhat 5.5 having 3 GB physical memory.

Possibly unrelated, but I’d compact that. Unless you really have a
working set of that size, or historical data about chef’s state is
that valuable to you.

Jonathan

Jonathan Matthews
London, UK
http://www.jpluscplusm.com/contact.html


#3

Which versions of CouchDB and Chef are you running? The Couch release
packaged for CentOS 5 (0.11.5, I think?) is quite ancient and known to have
such issues. If you’re having this problem in a post-1.0 release, that would
be more interesting.

That said – I’ve had a crashing CouchDB (pre-1.0), and no component other
than CouchDB itself has needed restart. (Further, if you use the
bootstrap-server recipe to deploy your server, Couch – like other
components – will be under process supervision and automatically
restarted).

On Thu, Feb 24, 2011 at 2:43 AM, le.huy@ingdirect.es wrote:

Hi all

I have experienced multi crashes of couchdb on my chef-server. I dont know
why, there is no log, no trace no indication from couchdb.
My database is around 30 GB and the server is redhat 5.5 having 3 GB
physical memory.

I heard that erlang sometime kill processes dueto lack of resource or other
contention, and restart them automatically. But in my case
the couchdb process (mochiweb I think) that faciliates database access via
http simply dies and I have restart couchdb entirely.

But the problem from Chef side is that Chef server can not re-connect to
couchdb again and I have to restart chef-server-api then chef-server-webui

Anyone is facing similar problem?

**

**

*Huy Le
**Tecnología **ING DIRECT
**Administrador de sistema
*Tel: +34 91 206 76 85

Por favor, no imprima este correo si no es necesario.


Attention:
The information contained in this message and or attachments is intended
only for the person or entity to which it is addressed and may contain
confidential and/or privileged material. Any review, retransmission,
dissemination or other use of, or taking of any action in reliance upon,
this information by persons or entities other than the intended recipient is
prohibited. If you received this in error, please contact the sender and
delete the material from any system and destroy any copies.


#4

+1 to this - make sure you are running a recent CouchDB, and make sure you
are doing regular compaction.

Adam

On Thu, Feb 24, 2011 at 7:06 AM, Charles Duffy charles@dyfis.net wrote:

Which versions of CouchDB and Chef are you running? The Couch release
packaged for CentOS 5 (0.11.5, I think?) is quite ancient and known to have
such issues. If you’re having this problem in a post-1.0 release, that would
be more interesting.

That said – I’ve had a crashing CouchDB (pre-1.0), and no component other
than CouchDB itself has needed restart. (Further, if you use the
bootstrap-server recipe to deploy your server, Couch – like other
components – will be under process supervision and automatically
restarted).

On Thu, Feb 24, 2011 at 2:43 AM, le.huy@ingdirect.es wrote:

Hi all

I have experienced multi crashes of couchdb on my chef-server. I dont
know why, there is no log, no trace no indication from couchdb.
My database is around 30 GB and the server is redhat 5.5 having 3 GB
physical memory.

I heard that erlang sometime kill processes dueto lack of resource or
other contention, and restart them automatically. But in my case
the couchdb process (mochiweb I think) that faciliates database access via
http simply dies and I have restart couchdb entirely.

But the problem from Chef side is that Chef server can not re-connect to
couchdb again and I have to restart chef-server-api then chef-server-webui

Anyone is facing similar problem?

**

**

*Huy Le
**Tecnología **ING DIRECT
**Administrador de sistema
*Tel: +34 91 206 76 85

Por favor, no imprima este correo si no es necesario.


Attention:
The information contained in this message and or attachments is intended
only for the person or entity to which it is addressed and may contain
confidential and/or privileged material. Any review, retransmission,
dissemination or other use of, or taking of any action in reliance upon,
this information by persons or entities other than the intended recipient is
prohibited. If you received this in error, please contact the sender and
delete the material from any system and destroy any copies.


Opscode, Inc.
Adam Jacob, Chief Product Officer
T: (206) 619-7151 E: adam@opscode.com