Random SystemExit. ActiveRecord::StatementInvalid

I have this recurrent problem that I just can’t figure out. On several
parts of the application when active record is executing some query it
throws a ActiveRecord::StatementInvalid exit like this:

A ActiveRecord::StatementInvalid occurred in
someapplication#some_action:

SystemExit: exit: SELECT * FROM project WHERE (project.id = 99999)
/usr/lib/ruby/gems/1.8/gems/activerecord-1.15.3/lib/active_record/connection_adapters/abstract_adapter.rb:128:in
`log’

The thing is that the next time you try the same action, it works fine,
also if you run that query directly in a console or any query browser it
returns a valid result. The occurrence is very random.

Any help or comments are greatly appreciated.

On Oct 28, 7:21 pm, Carlos G. [email protected]
wrote:

I have this recurrent problem that I just can’t figure out. On several
parts of the application when active record is executing some query it
throws a ActiveRecord::StatementInvalid exit like this:

I may be misremembering, but I have a vague idea that this can happen
if the mongrel/whatever is sent a TERM signal. How is your app
deployed? (in particular I seem to remember apache being a little
enthusiastic about killing fastcgi instances)

Fred

Frederick C. wrote:

On Oct 28, 7:21�pm, Carlos G. [email protected]
wrote:

I have this recurrent problem that I just can’t figure out. On several
parts of the application when active record is executing some query it
throws a ActiveRecord::StatementInvalid exit like this:

I may be misremembering, but I have a vague idea that this can happen
if the mongrel/whatever is sent a TERM signal. How is your app
deployed? (in particular I seem to remember apache being a little
enthusiastic about killing fastcgi instances)

Fred

We deploy to virtual hosts using apache on ubuntu.

On 28 Oct 2008, at 20:22, Carlos G.
<[email protected]

wrote:

I may be misremembering, but I have a vague idea that this can happen
if the mongrel/whatever is sent a TERM signal. How is your app
deployed? (in particular I seem to remember apache being a little
enthusiastic about killing fastcgi instances)

Fred

We deploy to virtual hosts using apache on ubuntu.

But what sits behind that? Mod_rails, mongrel, fastcgi, scgi?

On Oct 28, 9:22 pm, Carlos G. [email protected]
wrote:

Fred

We deploy to virtual hosts using apache on ubuntu.

But what sits behind that? Mod_rails, mongrel, fastcgi, scgi?

hehe, sorry, fastcgi

So as I was saying, I remember apache killing off fastcgi instances. I
think it logs (or can be told to log) when it does this. If you can
correlate this with when you get the errors then you’ll have found
your cause.

Fred

Frederick C. wrote:

On Oct 28, 9:22�pm, Carlos G. [email protected]
wrote:

Fred

We deploy to virtual hosts using apache on ubuntu.

But what sits behind that? Mod_rails, mongrel, fastcgi, scgi?

hehe, sorry, fastcgi

So as I was saying, I remember apache killing off fastcgi instances. I
think it logs (or can be told to log) when it does this. If you can
correlate this with when you get the errors then you’ll have found
your cause.

Fred

Hi,

I have the same problem… I am planning to speed up the query. Will it
help??

Regards
Selva

Frederick C. wrote:

On 28 Oct 2008, at 20:22, Carlos G.
<[email protected]

wrote:

I may be misremembering, but I have a vague idea that this can happen
if the mongrel/whatever is sent a TERM signal. How is your app
deployed? (in particular I seem to remember apache being a little
enthusiastic about killing fastcgi instances)

Fred

We deploy to virtual hosts using apache on ubuntu.

But what sits behind that? Mod_rails, mongrel, fastcgi, scgi?

hehe, sorry, fastcgi