diff options
author | Ross Brattain <ross.b.brattain@intel.com> | 2017-09-28 00:10:43 -0700 |
---|---|---|
committer | Ross Brattain <ross.b.brattain@intel.com> | 2017-10-04 08:31:17 -0700 |
commit | 8e35dc2c45ec1897690173b374406a05b2132954 (patch) | |
tree | e21fbaf51dd5d3aafe801798f5efa3e83be4b50d /yardstick/network_services/vnf_generic/vnf/prox_vnf.py | |
parent | ddff9781c46fc446380a378cbca40d9576e32b68 (diff) |
NSB PROX test hang fixes
The PROX tests were hanging in the duration
runner.
These are fixes for various errors:
raise error in collect_kpi if VNF is down
move prox dpdk_rebind after collectd stop
fix dpdk nicbind rebind to group by drivers
prox: raise error in collect_kpi if the VNF is down
prox: add VNF_TYPE for consistency
sample_vnf: debug and fix kill_vnf
pkill is not matching some executable names,
add some debug process dumps and try switching
back to killall until we can find the issue
sample_vnf: add default timeout, so we can override
default 3600 SSH timeout
collect_kpi is the point at which we check
the VNFs and TGs for failures or exits
queues are the problem make sure we aren't silently blocking on
non-empty queues by canceling join thread in subprocess
fixup duration runner to close queues
and other attempt to stop duration runner
from hanging
VnfdHelper: memoize port_num
resource: fail if ssh can't connect
at the end of 3600 second test our ssh connection
is dead, so we can't actually stop collectd
unless we reconnect
fix stop() logic to ignore ssh errors
Change-Id: I6c8e682a80cb9d00362e2fef4a46df080f304e55
Signed-off-by: Ross Brattain <ross.b.brattain@intel.com>
Diffstat (limited to 'yardstick/network_services/vnf_generic/vnf/prox_vnf.py')
-rw-r--r-- | yardstick/network_services/vnf_generic/vnf/prox_vnf.py | 41 |
1 files changed, 29 insertions, 12 deletions
diff --git a/yardstick/network_services/vnf_generic/vnf/prox_vnf.py b/yardstick/network_services/vnf_generic/vnf/prox_vnf.py index 2ac6ea412..3bfca19aa 100644 --- a/yardstick/network_services/vnf_generic/vnf/prox_vnf.py +++ b/yardstick/network_services/vnf_generic/vnf/prox_vnf.py @@ -16,9 +16,10 @@ import errno import logging +from yardstick.common.process import check_if_process_failed from yardstick.network_services.vnf_generic.vnf.prox_helpers import ProxDpdkVnfSetupEnvHelper from yardstick.network_services.vnf_generic.vnf.prox_helpers import ProxResourceHelper -from yardstick.network_services.vnf_generic.vnf.sample_vnf import SampleVNF +from yardstick.network_services.vnf_generic.vnf.sample_vnf import SampleVNF, PROCESS_JOIN_TIMEOUT LOG = logging.getLogger(__name__) @@ -51,10 +52,18 @@ class ProxApproxVnf(SampleVNF): try: return self.resource_helper.execute(cmd, *args, **kwargs) except OSError as e: - if not ignore_errors or e.errno not in {errno.EPIPE, errno.ESHUTDOWN}: + if e.errno in {errno.EPIPE, errno.ESHUTDOWN, errno.ECONNRESET}: + if ignore_errors: + LOG.debug("ignoring vnf_execute exception %s for command %s", e, cmd) + else: + raise + else: raise def collect_kpi(self): + # we can't get KPIs if the VNF is down + check_if_process_failed(self._vnf_process) + if self.resource_helper is None: result = { "packets_in": 0, @@ -64,12 +73,13 @@ class ProxApproxVnf(SampleVNF): } return result - intf_count = len(self.vnfd_helper.interfaces) - if intf_count not in {1, 2, 4}: + # use all_ports so we only use ports matched in topology + port_count = len(self.vnfd_helper.port_pairs.all_ports) + if port_count not in {1, 2, 4}: raise RuntimeError("Failed ..Invalid no of ports .. " "1, 2 or 4 ports only supported at this time") - port_stats = self.vnf_execute('port_stats', range(intf_count)) + port_stats = self.vnf_execute('port_stats', range(port_count)) try: rx_total = port_stats[6] tx_total = port_stats[7] @@ -94,13 +104,20 @@ class ProxApproxVnf(SampleVNF): self.setup_helper.tear_down() def terminate(self): + # stop collectd first or we get pika errors? + self.resource_helper.stop_collect() # try to quit with socket commands - self.vnf_execute("stop_all") - self.vnf_execute("quit") - # hopefully quit succeeds and socket closes, so ignore force_quit socket errors - self.vnf_execute("force_quit", _ignore_errors=True) - if self._vnf_process: - self._vnf_process.terminate() + # pkill is not matching, debug with pgrep + self.ssh_helper.execute("sudo pgrep -lax %s" % self.setup_helper.APP_NAME) + self.ssh_helper.execute("sudo ps aux | grep -i %s" % self.setup_helper.APP_NAME) + if self._vnf_process.is_alive(): + self.vnf_execute("stop_all") + self.vnf_execute("quit") + # hopefully quit succeeds and socket closes, so ignore force_quit socket errors + self.vnf_execute("force_quit", _ignore_errors=True) self.setup_helper.kill_vnf() self._tear_down() - self.resource_helper.stop_collect() + if self._vnf_process is not None: + LOG.debug("joining before terminate %s", self._vnf_process.name) + self._vnf_process.join(PROCESS_JOIN_TIMEOUT) + self._vnf_process.terminate() |