Bug 76631

Summary: "Lost connection to MySQL server during query" error since updating glibc
Product: [Retired] Red Hat Linux Reporter: Neil Sedley <nsedley>
Component: mysqlAssignee: Patrick Macdonald <patrickm>
Status: CLOSED DUPLICATE QA Contact: David Lawrence <dkl>
Severity: high Docs Contact:
Priority: medium    
Version: 7.1CC: bsherman, edoutreleau, ksnider, kth, tommi
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2002-11-07 01:43:52 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Neil Sedley 2002-10-24 12:13:55 UTC
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0; .NET CLR 
1.0.3705)

Description of problem:
Since updating glibc (via up2date) I get a "Lost connection to MySQL server 
during query" when trying to connect via an external IP (localhost is ok). 

Version-Release number of selected component (if applicable):


How reproducible:
Always

Steps to Reproduce:
1. mysql -p -h<external ip>
2. enter password

the above from another hosts sometimes works and sometimes crashes
	

Actual Results:  "Lost connection to MySQL server during query" error and 
return to command prompt

This error is returned using mysql and mysqladmin and occurs before any query 
takes place. When this occurs the mysql server crashes and restarts with the 
following log message

Number of processes running now: 1
mysqld process hanging, pid 7258 - killed
021021 16:24:11  mysqld restarted
/usr/libexec/mysqld: ready for connections


Expected Results:  I expect to get a mySQL command prompt.

Additional info:

RPMs installed

mysql-3.23.41-1
mysql-devel-3.23.41-1
mysqlclient9-3.23.22-6
mysql-server-3.23.41-1

Comment 1 Tomas Edwardsson 2002-10-28 15:02:41 UTC
This is also a problem with RedHat 8.0, no connections using IP numbers other 
than 127.0.0.1 work. If I try -h<IP of machine> it will fail and the following 
echoed into mysqld.log: 
 
Number of processes running now: 1 
mysqld process hanging, pid 15240 - killed 
021028 15:01:21  mysqld restarted 
 
I tried turning of this check alone in safe_mysqld and all that did for me was 
leaving the client session hanging.

Comment 2 Ken Snider 2002-10-29 22:44:16 UTC
I can verify this as well.

As of our update to glibc 2.2.4-30, any DNS query will kill mysql.

Placing the offending IP in /etc/hosts *does* work as a temporary workaround,
but it appears any attempt by mySQL to do a DNS query kills the server.

Comment 3 Need Real Name 2002-11-05 09:23:50 UTC
problem exists on RedHat 7.3 as well. Haven't check redhat 7.2

Comment 4 Eric Doutreleau 2002-11-05 16:02:13 UTC
it appears that the nscd daemon is dead.

when i manually restart the nscd daemon it work again without
the /etc/hosts until the nscd daemon crashed

Comment 5 Ken Snider 2002-11-05 16:51:54 UTC
It's not nscd specific - our systems don't run with nscd yet we see this problem. 

This affects any of our boxen (RH 7.1, 7.2, 7.3) using mysql if mysql needs to
access DNS for any reason. If we do not add the ip to /etc/hosts, mysql will crash.

Comment 6 Need Real Name 2002-11-07 01:43:42 UTC
I believe this is a duplicate of Bug 75128 for which an Errata was released today.
http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=75128

Comment 7 Neil Sedley 2002-11-07 08:15:06 UTC

*** This bug has been marked as a duplicate of 75128 ***