Date: 2010-10-19 08:41:56 +0200
From: @drstmane
To: SQL devs <>
Version: 2.40.1 (Oct2010) [obsolete]

Last updated: 2010-11-15 09:38:57 +0100

Comment 15065

Date: 2010-10-19 08:41:56 +0200
From: @drstmane

On Windows, several (currently 10 of 43) SQL tests that use pyhton timeout (i.e., potentially hang) on Windows, only; cf.,

http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_BugTracker-2009/create_on_ro_db_crash.SF-2830238.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_BugTracker-2009/lost_update.SF-2790020.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_BugTracker/connect_crash.SF-1436626.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_BugTracker/mdb_starts_with_sql_debug_64.SF-1999354.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_BugTracker/set_a_new_user_password.SF-1844050.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_BugTracker/set_sql_debug_64__breaking_the_DB.SF-1906287.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_Connections/connections.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_mserver5-sql-readonly/mserver5-sql-readonly-table.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_mserver5-sql-readonly/mserver5-sql-readonly-view.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Candidate/sql/.mTests103/Int.32.32.d.1-Windows5.1/src_test_VOC/VOC.err.00.html

Comment 15112

Date: 2010-10-26 17:36:14 +0200
From: @sjoerdmullender

Changeset 073b68dc838b made by Sjoerd Mullender sjoerd@acm.org in the MonetDB repo, refers to this bug.

For complete details, see http//devmonetdborg/hg/MonetDB?cmd=changeset;node=073b68dc838b

Changeset description:

Avoid deadlock.
This should fix bug #2699.
The deadlock happens especially on Windows where pipe buffers are
really small.  The server was hanging on writing to stderr and hence
not reading from stdin where we sent a command that was supposed to
produce output on stdout, which we were attempting to read.