On 03/15/2017 03:27 PM, John Ferlan wrote: > Fix a "bug" in the storage pool test driver code which "assumed" > testStoragePoolObjSetDefaults should fill in the configFile for > both the Define/Create (persistent) and CreateXML (transient) pools > by just VIR_FREE()'ing it during CreateXML. Because the configFile > was filled in, during Destroy the pool wouldn't be free'd which > could cause issues for future patches which add tests to validate > vHBA creation for the storage pool using the same name. > > Signed-off-by: John Ferlan <jferlan@xxxxxxxxxx> > --- > src/test/test_driver.c | 6 ++++++ > 1 file changed, 6 insertions(+) > > diff --git a/src/test/test_driver.c b/src/test/test_driver.c > index cf7820a..361d62e 100644 > --- a/src/test/test_driver.c > +++ b/src/test/test_driver.c > @@ -4440,6 +4440,12 @@ testStoragePoolCreateXML(virConnectPtr conn, > pool = NULL; > goto cleanup; > } > + > + /* *SetDefaults fills this in for the persistent pools, but this > + * would be a transient pool so remove it; otherwise, the Destroy > + * code will not Remove the pool */ > + VIR_FREE(pool->configFile); > + > pool->active = 1; > > event = virStoragePoolEventLifecycleNew(pool->def->name, pool->def->uuid, > After looking at the context, this makes sense. ACK. -- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list
