tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bugzi...@apache.org
Subject DO NOT REPLY [Bug 32320] New: - Make mod_jk lb retry interval configurable
Date Fri, 19 Nov 2004 13:10:50 GMT
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG·
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=32320>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND·
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=32320

           Summary: Make mod_jk lb retry interval configurable
           Product: Tomcat 5
           Version: 5.0.30
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: enhancement
          Priority: P2
         Component: Native:JK
        AssignedTo: tomcat-dev@jakarta.apache.org
        ReportedBy: rainer.jung@kippdata.de


Make mod_jk lb retry interval configurable

In mod_jk 1.2.7 inside common/jk_lb_worker.c there are constants:

/* 
 * Time to wait before retry...
 */
#define WAIT_BEFORE_RECOVER (60*1)
#define ADDITINAL_WAIT_LOAD (20)

The second constant ADDITINAL_WAIT_LOAD can be deleted, because it is no longer
used (since Mladen Turks new loda balancing algorithm).

The first constant WAIT_BEFORE_RECOVER should be made an configurable attribute
of the corresponding lb worker. Since this constant determines how often a
failed worker is retried and every failed retry produces error log lines (with
apache 1.3 for every process) and decrements the lb_value, it should be made
user adjustable to decide, after which intervals retries should be done.

Additionally one could make the retry interval for any worker a little more
dynamic: Use the configured interval the first time after a worker failed, the
next two failed retries will double the interval. Then keep it constant and
reset after successful retry.

Both changes are easy to do, and I would be willing to provide patches.

With the momentary setup using apache 1.3 with e.g. 50 processes under load you
will see 50 error log blocks per minute, so nearly one block per second.

-- 
Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

---------------------------------------------------------------------
To unsubscribe, e-mail: tomcat-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tomcat-dev-help@jakarta.apache.org


Mime
View raw message