[SSSD] [PATCH] SDAP: augmented logging for group saving
Jakub Hrozek
jhrozek at redhat.com
Thu Apr 17 15:30:45 UTC 2014
On Tue, Apr 15, 2014 at 04:25:50PM +0200, Pavel Reichl wrote:
> Hello,
>
> some more generally useful logging messages are needed for that
> bugzilla.
>
> Patches for both master and 1.11 are attached.
>
> Thanks for review!
It would be nice to explain in the commit message why you used TRACE_LIBS
for the messages and not i.e OP_FAILURE.
>
> On Wed, 2014-04-09 at 14:29 +0200, Pavel Reichl wrote:
> > Hello,
> >
> > please see attached patch.
> >
> > This patch was previously written for BZ 1059423. But it now seems that
> > more detailed logging information is generally useful for issues that
> > are emerging from this area lately.
> >
> > Pavel Reichl
>
>
> From e7e4a06098bce4df600370274babe470b0d71c47 Mon Sep 17 00:00:00 2001
> From: Pavel Reichl <reichl.pavel at gmail.com>
> Date: Mon, 14 Apr 2014 17:27:23 +0100
> Subject: [PATCH] SDAP: augmented logging when adding new group
>
> Resolves:
> https://fedorahosted.org/sssd/ticket/2239
> ---
> src/db/sysdb_ops.c | 92 +++++++++++++++++++++++++++++-----
> src/providers/ldap/sdap_async_groups.c | 2 +-
> 2 files changed, 81 insertions(+), 13 deletions(-)
>
> diff --git a/src/db/sysdb_ops.c b/src/db/sysdb_ops.c
> index 95354d98806a13cf142705dc304b7810b8ee82f4..559f8b51813fa499ec21d9e78189910cacc7b32a 100644
> --- a/src/db/sysdb_ops.c
> +++ b/src/db/sysdb_ops.c
> @@ -1428,7 +1428,15 @@ int sysdb_add_group(struct sysdb_ctx *sysdb,
> ret = sysdb_search_user_by_name(tmp_ctx, sysdb, domain,
> name, NULL, &msg);
> if (ret != ENOENT) {
> - if (ret == EOK) ret = EEXIST;
> + if (ret == EOK) {
> + DEBUG(SSSDBG_TRACE_LIBS, ("MPG domain is having a user "
"...domain contains..."
> + "with the same name - %s.\n", name));
> + ret = EEXIST;
> + } else {
> + DEBUG(SSSDBG_TRACE_LIBS,
> + ("sysdb_search_user_by_name failed for user %s.\n",
> + name));
> + }
> goto done;
> }
> }
[snip]
> goto done;
> }
> DEBUG(SSSDBG_MINOR_FAILURE,
> ("A group with the same GID [%llu] was removed from the "
> "cache\n", (unsigned long long) gid));
If you're touching the code anyway, can you change the DEBUG calls to
use SPRIgid here as well?
> +
> ret = sysdb_add_group(sysdb, domain, name, gid,
> attrs, cache_timeout, now);
> + if (ret) {
> + DEBUG(SSSDBG_MINOR_FAILURE,
> + ("sysdb_add_group failed (while renaming group) for: "
> + "%s [%"SPRIgid"].\n", name, gid));
> + }
> }
> goto done;
> }
More information about the sssd-devel
mailing list