[PATCH] Fix crash and/or error when diskpool.conf does not exist

# HG changeset patch # User Dan Smith <danms@us.ibm.com> # Date 1195073495 28800 # Node ID 7798be3698b4a53c1301dc3e3f4a4bfd1bc9e37b # Parent d2bcdde08df6080caa9ae8d152d516d232fb6345 Fix crash and/or error when diskpool.conf does not exist Previously, a missing diskpool configuration file returned an error, which prevented enumeration of all CIM_ResourcePool objects. This changes that behavior to consider a lack of a diskpool.conf file to mean "no disk pools have been defined", equivalent to how a lack of bridges indicates that no network pools have been defined. Signed-off-by: Dan Smith <danms@us.ibm.com> diff -r d2bcdde08df6 -r 7798be3698b4 src/Virt_DevicePool.c --- a/src/Virt_DevicePool.c Tue Nov 13 14:18:17 2007 -0800 +++ b/src/Virt_DevicePool.c Wed Nov 14 12:51:35 2007 -0800 @@ -505,18 +505,14 @@ static CMPIStatus diskpool_instance(virC const char *id, const CMPIBroker *broker) { - CMPIStatus s; + CMPIStatus s = {CMPI_RC_OK, NULL}; struct disk_pool *pools = NULL; int count = 0; int i; count = get_diskpool_config(&pools); - if ((id == NULL) && (count == 0)) { - cu_statusf(_BROKER, &s, - CMPI_RC_ERR_FAILED, - "No such pool `%s'", id); - goto out; - } + if ((id == NULL) && (count == 0)) + return s; for (i = 0; i < count; i++) { CMPIInstance *pool; @@ -533,8 +529,6 @@ static CMPIStatus diskpool_instance(virC inst_list_add(list, pool); } - CMSetStatus(&s, CMPI_RC_OK); - out: free_diskpool(pools, count); return s;

Dan Smith wrote:
# HG changeset patch # User Dan Smith <danms@us.ibm.com> # Date 1195073495 28800 # Node ID 7798be3698b4a53c1301dc3e3f4a4bfd1bc9e37b # Parent d2bcdde08df6080caa9ae8d152d516d232fb6345 Fix crash and/or error when diskpool.conf does not exist
Previously, a missing diskpool configuration file returned an error, which prevented enumeration of all CIM_ResourcePool objects. This changes that behavior to consider a lack of a diskpool.conf file to mean "no disk pools have been defined", equivalent to how a lack of bridges indicates that no network pools have been defined.
Signed-off-by: Dan Smith <danms@us.ibm.com>
Seems like a reasonable change. I ran into this issue last week. =) +1 -- Kaitlin Rupert IBM Linux Technology Center karupert@us.ibm.com
participants (2)
-
Dan Smith
-
Kaitlin Rupert