Red Hat Bugzilla – Bug 76578
connect using mysql_read_default_file from perl scripts fails
Last modified: 2007-04-18 12:47:54 EDT
From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.1) Gecko/20020826 Description of problem: Connecting to mysql server using DBI methods as described in http://www.mysql.com/documentation/mysql/bychapter/manual_Clients.html#IDX1534 fails if mysql_read_default_file is used. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Use mysql_read_default_file as described in http://www.mysql.com/documentation/mysql/bychapter/manual_Clients.html#IDX1534 2. 3. Actual Results: Apache Log gives 'premature end of script header'. Webpage shows 'Internal server error' Expected Results: Webpage with database query results Additional info: The bug appeared somewhere around mysql 3.23.49 and is fixed in mysql 3.23.50 or later. Replacing libmysqlclient.so.10.0.0 in RedHat 7.3 (251027 bytes) with the version from RedHat 7.2 (266812 bytes) fixes the problem (pre bug version?).
I was not able to reproduce this error using the current version of MySQL (3.23.54a) available for RHL 7.3. This is consistent with your findings as listed under "Additional Information". Please upgrade to the lastest available version of MySQL if you have not done so already. Thanks for letting us know about the problem.
---- Additional Comments From mranweil@us.ibm.com 2005-07-26 13:08 EDT ------- tar of logs, plus a dmesg from a different attempt Sorry, for some reason I thought I had attached (but obviously hadn't). The dmesg is from a different install (trying default partitioning from that disk rather than manual). The tmpsdd and devsdd are basically just listings of sdd* in the appropritate (/dev or /tmp) directory. It's the node in /tmp that is mounted during the install, even though fdisk doesn't show sdd1 existing.