Summary: | dev-db/mysql-init-scripts broken with baselayout-1.13.0 | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Bjarke Istrup Pedersen (RETIRED) <gurligebis> |
Component: | Current packages | Assignee: | Gentoo Linux MySQL bugs team <mysql-bugs> |
Status: | RESOLVED DUPLICATE | ||
Severity: | normal | CC: | hollow, jakub |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | 158769 | ||
Bug Blocks: |
Description
Bjarke Istrup Pedersen (RETIRED)
2007-02-13 02:24:03 UTC
Revert changes to fix bug #158769 sed -e 's|#see bug #158769||' -i /etc/init.d/mysql worksforme, a better fix should be provided though, I'm available to consulting for everyone interested in fix/rewrite the mysql-init-scripts package, just email or query me on irc (In reply to comment #1) > Revert changes to fix bug #158769 > sed -e 's|#see bug #158769||' -i /etc/init.d/mysql > worksforme, a better fix should be provided though, I'm available to consulting > for everyone interested in fix/rewrite the mysql-init-scripts package, just > email or query me on irc That doesn't fix it here, tried it, and the output is still the same :-( workarounded in "=dev-db/mysql-init-scripts-1.2", there is a race condition in the code, still not catched. wondering if http://bugs.mysql.com/bug.php?id=21884 is related to this Removing "--pidfile ${pidfile}" from the end of line 283 of /etc/init.d/mysql fixes the problem for me. The start-stop-daemon seems to be creating the pid file, which mysql then overwrites with it's own pid file, and somewhere in there... I don't know, a race condition maybe... that results in the server being up, but the init script thinking it failed. Hard to pinpoint, as it only happens on boot, not regular start/stop, and I've got a headless server. As a side-note, I think the "DEBUG" setting in conf.d/mysql should actually result in debug info going into the logs, not just the console, for exactly this sort of case. Packages: dev-db/mysql-5.0.38 dev-db/mysql-init-scripts-1.2 sys-apps/baselayout-1.13.0_alpha12 what's the status here? still doesn't work with baselayout-2.. mysql starts, but output is still borked: * Starting ... * Starting (/etc/mysql/my.cnf) [ ok ] ..... * Stopping ... * Stopping mysqld (0) [ ok ] Since the mysql init script is such a fcking mess i won't touch it, so please fix it asap... (In reply to comment #6) > what's the status here? still doesn't work with baselayout-2.. mysql starts, > but output is still borked: baselayout-2 is package.masked, for starters (presumably for a reason). If someone would create some transition documentation for ebuild maintainers (as baselayout/s-s-d updates keep breaking initscripts over and over again), life would be much easier. Problem still persists with a brand-new mysql installation on a vserver guest maseltov ~ # vemerge abdullah -- -pv baselayout mysql These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] sys-apps/baselayout-1.13.0_alpha12 USE="pam unicode -bootstrap -build -static" 226 kB [ebuild R ] dev-db/mysql-5.0.42 USE="berkdb perl ssl -big-tables -cluster -debug -embedded -extraengine -latin1 -max-idx-128 -minimal (-selinux) -static" 0 kB Solution from Xepher (comment #5) worked for me, too. Best, Simon This is where working init scripts are *** This bug has been marked as a duplicate of bug 175783 *** |