On Fri, Mar 1, 2013 at 5:44 AM, Ian Kent <raven@xxxxxxxxxx> wrote:
> On Wed, 2013-02-27 at 22:49 -0300, Leonardo Chiquitto wrote:
>> Hello,
>>
>> I've got a bug report describing a case where AutoFS fails to read new
>> entries from /etc/auto.master after a SIGHUP. Although the problem
>> was reported in an older version of automount, it is reproducible using
>> the latest revision from git.
>>
>> How to reproduce:
>>
>> 1. Add at least two sources of AutoFS maps to /etc/nsswitch.conf. I've
>> tested only with "files nis". You don't need to configure NIS/YP, just
>> having it listed there in the configuration is enough.
>>
>> 2. Start the automounter with a simple /etc/auto.master:
>>
>> /nfs1 /etc/auto.test1
>> +auto.master
>>
>> 3. Add another entry to /etc/auto.master:
>>
>> /nfs1 /etc/auto.test1
>> /nfs2 /etc/auto.test2
>> +auto.master
>>
>> and reload the daemon. Notice that although AutoFS reads /etc/auto.test2,
>> /nfs2 is not created/mounted.
>>
>> 4. Try to stop the daemon cleanly (SIGTERM only). You'll notice that it won't
>> quit. SIGKILL is necessary.
>
> While the failure to exit when this occurs is a worry probably due to
> the presence of a master map entry in the list that was not acted upon.
> We probably should come up with way to produce this problem after the
> re-read problem is fixed so we can check what is actually happening.
Understood. I'll try to write a patch to make it ignore these "partial" entries
when exiting.
> In the mean time have a look at this patch.
>
> autofs-5.0.7 - fix map read fail incorrectly set on master re-read
I tried it here and unfortunately I'm still able to reproduce the problem.
I haven't had time to debug it further yet, but my *impression* is that
when reading the entry "+auto.master" from /etc/auto.master it
fails with nss source "files" due to the recursion and sets read_fail to 1.
Next it tries with nss source "nis" and succeeds, lookup_nss_read_master()
returns SUCCESS and the "read_fail = 0" added to lookup_read_master()
is never executed.
Have a nice weekend!
Leonardo
> From: Ian Kent <raven@xxxxxxxxxx>
>
> When re-reading the master map read failures for plus included maps
> should not cause the map read to fail, they should be ingored.
> ---
> CHANGELOG | 1 +
> modules/lookup_file.c | 7 ++++++-
> 2 files changed, 7 insertions(+), 1 deletion(-)
>
> diff --git a/CHANGELOG b/CHANGELOG
> index 39388a5..b13a5f5 100644
> --- a/CHANGELOG
> +++ b/CHANGELOG
> @@ -28,6 +28,7 @@
> - make yellow pages support optional.
> - modules/replicated.c: use sin6_addr.s6_addr32.
> - workaround missing GNU versionsort extension.
> +- fix map read fail incorrectly set on master re-read.
>
> 25/07/2012 autofs-5.0.7
> =======================
> diff --git a/modules/lookup_file.c b/modules/lookup_file.c
> index facb305..bca92fd 100644
> --- a/modules/lookup_file.c
> +++ b/modules/lookup_file.c
> @@ -443,7 +443,7 @@ int lookup_read_master(struct master *master, time_t age, void *context)
>
> inc = check_master_self_include(master, ctxt);
> if (inc)
> - master->recurse = 1;;
> + master->recurse = 1;
> master->depth++;
> status = lookup_nss_read_master(master, age);
> if (!status) {
> @@ -451,6 +451,11 @@ int lookup_read_master(struct master *master, time_t age, void *context)
> MODPREFIX
> "failed to read included master map %s",
> master->name);
> + /*
> + * Plus map include failures don't cause the map
> + * read to fail.
> + */
> + master->read_fail = 0;
> }
> master->depth--;
> master->recurse = 0;
--
To unsubscribe from this list: send the line "unsubscribe autofs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
