Re: [libvirt] [PATCH 7/7] storage: Check for duplicate host for incoming pool def

On 04/20/2015 11:11 AM, Ján Tomko wrote: > On Sun, Apr 19, 2015 at 08:49:12PM -0400, John Ferlan wrote: >> https://bugzilla.redhat.com/show_bug.cgi?id=1171984 >> >> Use the new virIsSameHostnameInfo API to determine whether the proposed >> storage pool definition matches the existing storage pool definition >> >> Signed-off-by: John Ferlan <jferlan(a)redhat.com&gt; >> --- >> src/conf/storage_conf.c | 11 ++++++++++- >> 1 file changed, 10 insertions(+), 1 deletion(-) >> >> diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c >> index 4852dfb..c1bc242 100644 >> --- a/src/conf/storage_conf.c >> +++ b/src/conf/storage_conf.c >> @@ -2415,7 +2415,16 @@ virStoragePoolSourceMatchSingleHost(virStoragePoolSourcePtr poolsrc, >> if (poolsrc->hosts[0].port != defsrc->hosts[0].port) >> return false; >> > > This function is called when parsing the configuration, which should not > depend on host state. > > For example, if libvirt is started really early at boot time and the > hostnames cannot be resolved by the DNS yet, they will pass the check > but they will disappear on libvirtd restart. Hmm... the downside of unreliable dependencies.

> > The hostname->ip pairings are not stable either, so if we do this, I > think it should be done on pool startup, not config parsing. > Right, but by the time we get to pool startup we'd be at the same point as referenced above - if done early enough at boot time, then it's not going to fail, but perhaps better than nothing.