Terminating glusterfs client with a signal leaves a stale fuse mount. Prior to dynamic volume changes (79241696) cleanup_and_exit() recursed over the xlator graph calling their fini() methods. This is not happening now, therefore eg. fuse mount stays in place, defunct (other side effects might occur, this is just the one I observed).
*** This bug has been marked as a duplicate of bug 971 ***
PATCH: http://patches.gluster.com/patch/3462 in master (io-threads: initialize conf->w_attr)
Avati, (In reply to comment #2) > PATCH: http://patches.gluster.com/patch/3462 in master (io-threads: initialize > conf->w_attr) Did you maybe mistype the error number? This bug has nothing to do with either of io-threads or Solaris, therefore I guess so...
> > Did you maybe mistype the error number? > > This bug has nothing to do with either of io-threads or Solaris, therefore I > guess so... this was regarding io-threads on server side solaris used to hang the server process.
> this was regarding io-threads on server side solaris used to hang the server > process. Indeed the bug has no resemblance for the patch :)