On Wed, Mar 24, 2010 at 1:04 PM, Benoit Chesneau <bchesn...@gmail.com> wrote:
> On Wed, Mar 24, 2010 at 10:40 AM, Brian Candler <b.cand...@pobox.com> wrote:
>> Now installs fine under Ubuntu Hardy (8.04) i386, once I'd remembered to
>> rm -rf /usr/local/lib/couchdb/erlang/lib.
>>
>> However, the in-browser test suite gave me various intermittent errors.
>>
>> Quite consistently I could get this one, even after disabling firefox
>> extensions(*):
>>
>> reader_acl      error   190ms
>>   1. Exception raised: {}
>>
>> It repeated consistently when clicking the arrow to re-run it.
>>
>> Clicking "run with debugger" just seemed to freeze the browser, although
>> maybe that's because I had disabled firebug at this point.
>>
>> So I re-enabled firebug, restarted firefox, and tried re-running just the
>> readeracl test again.  This time, I managed to crash the couchdb server
>> itself!
>>
>> $ telnet localhost 5984
>> Trying 127.0.0.1...
>> telnet: Unable to connect to remote host: Connection refused
>>
>> Logs showed:
>>
>> [Wed, 24 Mar 2010 08:52:21 GMT] [info] [<0.5319.0>] 127.0.0.1 - - 'POST' 
>> /test_suite_users/_ensure_full_commit 201
>>
>> [Wed, 24 Mar 2010 08:52:21 GMT] [info] [<0.5319.0>] 127.0.0.1 - - 'POST' 
>> /_restart 200
>>
>> [Wed, 24 Mar 2010 08:52:21 GMT] [error] [<0.5301.0>] ** Generic server 
>> couch_config terminating
>> ** Last message in was {set,"couch_httpd_auth","secret",
>>                            "f8da3b2729ce9bea76220f692378eb0b",true}
>> ** When Server state == {config,[{<0.5349.0>,#Fun<couch_config.2.17946139>},
>>                                 {false,#Fun<couch_server.3.35387302>},
>>                                 {<0.5347.0>,#Fun<couch_config.3.70971031>},
>>                                 {<0.5347.0>,#Fun<couch_config.2.17946139>},
>>                                 {<0.5320.0>,#Fun<couch_config.2.17946139>},
>>                                 {<0.5318.0>,#Fun<couch_config.2.17946139>},
>>                                 {<0.5315.0>,#Fun<couch_config.2.17946139>},
>>                                 {<0.5315.0>,#Fun<couch_config.2.17946139>},
>>                                 {<0.5313.0>,#Fun<couch_config.3.70971031>},
>>                                 {<0.5312.0>,#Fun<couch_config.2.17946139>},
>>                                 {<0.5310.0>,#Fun<couch_config.2.17946139>},
>>                                 {false,#Fun<couch_server.3.35387302>},
>>                                 {<0.4.0>,#Fun<couch_config.2.17946139>}],
>>                                "/usr/local/etc/couchdb/local.ini"}
>> ** Reason for termination ==
>> ** {{badmatch,{error,eacces}},
>>    [{couch_config_writer,save_to_file,2},
>>     {couch_config,handle_call,3},
>>     {gen_server,handle_msg,5},
>>     {proc_lib,init_p_do_apply,3}]}
>>
>>
>> [Wed, 24 Mar 2010 08:52:21 GMT] [error] [<0.5301.0>] {error_report,<0.29.0>,
>>              {<0.5301.0>,crash_report,
>>               [[{pid,<0.5301.0>},
>>                 {registered_name,couch_config},
>>                 {error_info,{exit,{{badmatch,{error,eacces}},
>>                                    [{couch_config_writer,save_to_file,2},
>>                                     {couch_config,handle_call,3},
>>                                     {gen_server,handle_msg,5},
>>                                     {proc_lib,init_p_do_apply,3}]},
>>                                   [{gen_server,terminate,6},
>>                                    {proc_lib,init_p_do_apply,3}]}},
>>                 {initial_call,{couch_config,init,['Argument__1']}},
>>                 {ancestors,[couch_server_sup,<0.30.0>]},
>>                 {messages,[]},
>>                 {links,[<0.75.0>]},
>>                 {dictionary,[]},
>>                 {trap_exit,false},
>>                 {status,running},
>>                 {heap_size,6765},
>>                 {stack_size,23},
>>                 {reductions,42634}],
>>                []]}}
>>
>> [Wed, 24 Mar 2010 08:52:21 GMT] [error] [<0.75.0>] {error_report,<0.29.0>,
>>    {<0.75.0>,supervisor_report,
>>     [{supervisor,{local,couch_server_sup}},
>>      {errorContext,child_terminated},
>>      {reason,
>>          {{badmatch,{error,eacces}},
>>           [{couch_config_writer,save_to_file,2},
>>            {couch_config,handle_call,3},
>>            {gen_server,handle_msg,5},
>>            {proc_lib,init_p_do_apply,3}]}},
>>      {offender,
>>          [{pid,<0.5301.0>},
>>           {name,couch_config},
>>           {mfa,
>>               {couch_server_sup,couch_config_start_link_wrapper,
>>                   [["/usr/local/etc/couchdb/default.ini",
>>                     "/usr/local/etc/couchdb/local.ini",
>>                     "/usr/local/etc/couchdb/default.ini",
>>                     "/usr/local/etc/couchdb/local.ini"],
>>                    <0.74.0>]}},
>>           {restart_type,permanent},
>>           {shutdown,brutal_kill},
>>           {child_type,worker}]}]}}
>>
>> [Wed, 24 Mar 2010 08:52:21 GMT] [error] [<0.75.0>] {error_report,<0.29.0>,
>>    {<0.75.0>,supervisor_report,
>>     [{supervisor,{local,couch_server_sup}},
>>      {errorContext,shutdown},
>>      {reason,reached_max_restart_intensity},
>>      {offender,
>>          [{pid,<0.5301.0>},
>>           {name,couch_config},
>>           {mfa,
>>               {couch_server_sup,couch_config_start_link_wrapper,
>>                   [["/usr/local/etc/couchdb/default.ini",
>>                     "/usr/local/etc/couchdb/local.ini",
>>                     "/usr/local/etc/couchdb/default.ini",
>>                     "/usr/local/etc/couchdb/local.ini"],
>>                    <0.74.0>]}},
>>           {restart_type,permanent},
>>           {shutdown,brutal_kill},
>>           {child_type,worker}]}]}}
>>
>> [Wed, 24 Mar 2010 08:52:21 GMT] [error] [<0.5319.0>] Uncaught error in HTTP 
>> request: {exit,
>>                                 {{{badmatch,{error,eacces}},
>>                                   [{couch_config_writer,save_to_file,2},
>>                                    {couch_config,handle_call,3},
>>                                    {gen_server,handle_msg,5},
>>                                    {proc_lib,init_p_do_apply,3}]},
>>                                  {gen_server,call,
>>                                   [couch_config,
>>                                    {set,"couch_httpd_auth","secret",
>>                                     "f8da3b2729ce9bea76220f692378eb0b",
>>                                     true}]}}}
>>
>> [Wed, 24 Mar 2010 08:52:21 GMT] [error] [emulator] Error in process 
>> <0.5356.0> with exit value: 
>> {badarg,[{ets,insert,[stats_hit_table,{{couchdb,open_databases},-1}]},{couch_stats_collector,decrement,1}]}
>>
>>
>>
>> [Wed, 24 Mar 2010 08:52:21 GMT] [error] [emulator] Error in process 
>> <0.5352.0> with exit value: 
>> {badarg,[{ets,insert,[stats_hit_table,{{couchdb,open_os_files},-1}]},{couch_stats_collector,decrement,1}]}
>>
>> eaccess suggested a permissions problem, so I did
>>  chown couchdb:couchdb /usr/local/etc/couchdb/local.ini
>> and restarted couchdb. After that this test suite was OK. However it's
>> not clear to me why it hadn't crashed before.
>>
>> Before then I had also seen intermittently other tests failing with the same
>> error - whole groups of them (**).  For example, on one run everything from
>> reduce_builtin to view_pagination inclusive failing with the same "Exception
>> raised: {}" error.
>>
>> I checked the log after that run and found some odd errors:
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.113.0>] OS Process #Port<0.1864> 
>> Log :: function raised exception (new ReferenceError("gc is not defined", 
>> "")) with doc._id 8450bfd7a0371f80a9baf032553367f4
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.103.0>] checkpointing view update 
>> at seq 1 for test_suite_db _temp
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.94.0>] 127.0.0.1 - - 'POST' 
>> /test_suite_db/_temp_view 200
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.113.0>] OS Process #Port<0.1864> 
>> Log :: function raised exception (new ReferenceError("map_funs is not 
>> defined", "")) with doc._id 8450bfd7a0371f80a9baf032553367f4
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.115.0>] checkpointing view update 
>> at seq 1 for test_suite_db _temp
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.94.0>] 127.0.0.1 - - 'POST' 
>> /test_suite_db/_temp_view 200
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.113.0>] OS Process #Port<0.1864> 
>> Log :: function raised exception (new ReferenceError("map_results is not 
>> defined", "")) with doc._id 8450bfd7a0371f80a9baf032553367f4
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.125.0>] checkpointing view update 
>> at seq 1 for test_suite_db _temp
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.94.0>] 127.0.0.1 - - 'POST' 
>> /test_suite_db/_temp_view 200
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.125.0>] Shutting down view group 
>> server, monitored db is closing.
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.115.0>] Shutting down view group 
>> server, monitored db is closing.
>>
>> [Wed, 24 Mar 2010 09:00:50 GMT] [info] [<0.103.0>] Shutting down view group 
>> server, monitored db is closing.
>>
>> Anyway, nobody else seems to have had this problem, it's quite possibly
>> something to do with my setup, and browser-based tests are fragile anyway.
>> Hence I'm not going to vote against the release.  It's a 0 from me.
>>
>> Regards,
>>
>> Brian.
>>
>> (*) firebug, flashblock, jssh, open in browser, user agent switcher; leaving
>> only "Ubuntu Firefox Modifications 0.5"
>>
>> (**) On my first run of the full test suite, it failed at
>>
>> reader_acl      error   190ms
>>   1. Exception raised: {}
>>
>> The subsequent tests were OK from recreate_doc down to security_validation
>> (inclusive), but then all the tests from show_documents to view_xml
>> (inclusive) all failed with the same error:
>>
>> show_documents  error   13282ms
>>   1. Exception raised: {}
>>
>> stats   error   18ms
>>   1. Exception raised: {}
>>
>> update_documents        error   14ms
>>   1. Exception raised: {}
>> ... etc
>>
>> So I reloaded the page and ran all the tests from the top. On this second
>> run, tests from form_submit down to stats (inclusive) failed in the same way.
>>
> I reported the same yesterday. Had to remove the db folder and
> relaunch the tests. Something happend in last 2 weeks about these
> tests.
>
> - benoit
>

I can only reproduce that on my macbook anyway. Do we change something
in the way we delete/reset the db recently ? I may have missed a
commit.

- benoit

Reply via email to