On Mon, 2009-10-26 at 20:15 -0400, Simo Sorce wrote:
On Mon, 2009-10-26 at 18:56 -0400, Simo Sorce wrote:
> I was seeing in some non-easily reproduceable cases errors returned by
> ldap_search_ext() that would hose the connection. The same connection
> handler would be used after the offline timeout causing again the same
> error (and us going offline again).
> This patch should fix it or at least make the code a bit more resilient
> to this kind of errors.
Self NACK while temporarily investigating a segfault possibly related to
ctx->gsh sometimes not being a vaild pointer ... will respin later.
ok found the error, calling mark_offline from within mark_offline,
recursively was not a very bright idea :-)
Attached, new patch that calls the right function (be_mark_offline).
--
Simo Sorce * Red Hat, Inc * New York