Hi everyone,
I have been trying to solve this problem for quite some time now and I
really don’t have any further clue to investigate.
I am running a mongrel_cluster 1.0.5-2 (mongrel verison : 1.1.5-5) on a
ubuntu box with apache2 2.2.12-1ubuntu2.1
I have 4 different apps on the same box and each app has two mongrel
instances.
After a short period of time ( < 2 hours ) some of the 8 mongrel
instances
just stop responding.
I read the mongrel FAQ and tried to
set ActiveRecord::Base.verification_timeout = 14400 (the mysql parameter
is
set to the default 28800) in myapp/config/environments/production.rb
The problem kept happening.
I decided to lower that number, making it 60.
No more success.
I installed gdb and the ruby macros to make sure i was in the mysql
driver
while hanging and sure enough, here is the top of the BT from GDB :
/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/mysql_adapter.rb:264:in
active?'", "/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract_adapter.rb:136:in
verify!’",
“/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_pool.rb:256:in
checkout_and_verify'", "/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_pool.rb:252:in
checkout_existing_connection’”,
“/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_pool.rb:186:in
checkout'", "/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_pool.rb:184:in
loop’”,
“/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_pool.rb:184:in
checkout'", "/usr/lib/ruby/1.8/monitor.rb:242:in
synchronize’”,
“/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_pool.rb:183:in
checkout'", "/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_pool.rb:98:in
connection’”,
“/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_pool.rb:326:in
retrieve_connection'", "/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_specification.rb:121:in
retrieve_connection’”,
“/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/connection_adapters/abstract/connection_specification.rb:113:in
connection'", "/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/base.rb:2934:in
quoted_table_name’”,
“/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/base.rb:1626:in
construct_finder_sql'", "/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/base.rb:1490:in
find_every’”,
“/var/lib/gems/1.8/gems/activerecord-2.2.2/lib/active_record/base.rb:589:in
`find’”,
I figured that since I was running a packaged version of mongrel_cluster
(not the gem installed one), I should use the package version of mysql
driver for ruby.
So i did
gem uninstall mysql
and installed :
apt-get install libmysql-ruby1.8
I then restarted the mongrels and sure enough after 1h25 one of them was
hanging in the mysql_adapter.rb, showing the same backtrace as before.
I really am out of clue here and any advice / help would be much
appreciated.
I have to note that I am far from being an ROR guru and maybe the
solution
to my problem is simpler than what I have been looking for, but I just
don’t
know where to look anymore.
Thanks for any advice / answer
Damien