def load(cache, trials, sleep, timeout):
act = Action.GETCOMMPEERS
log.info("Start %s:..." % (act))
success = False
try:
for _ in range(trials):
try:
CacheCommPeer.__loadTry(cache.ui, cache)
success = True
break
except urllib2.HTTPError as inst:
cache.reconnect(sleep, timeout)
if not success:
log.error("exhausted trials; could not %s %s" % (act, cache))
except Exception as inst:
log.error("could not digest comm peer status %s (%s)" % (str(inst), traceback.format_exc()))
finally:
log.debug("Stop %s: Success %s..." % (act, success))
return success
def __loadTry(ui, auId):
'''
delete existing LockssAuCrawlStatus and create new by reading status info from cache
'''
log.debug("try %s" % (auId))
st = { 'auId' : auId }
status = ui.getCrawlStatus(auId.masterAuId.getLockssAu())
reportDate = datetime.utcnow().replace(tzinfo=pytz.utc)
if (not status):
log.debug2("No CrawlStatus Info for %s %s" % (auId.cache, auId.auId) )
else:
for s in status:
# work on status info
if (not s):
raise LockssError("CrawlStatusTable returned empty info");
try:
st['reportDate'] = reportDate
st['type'] = s['crawl_type']
st['startTime'] = datetime.strptime(s['start'], utils.UI_STRFTIME)
st['nBytesFetched'] = s['content_bytes_fetched'].replace(",", "")
st['status'] = s['crawl_status']['value']
st['nMimeTypes'] = str(s['num_of_mime_types']['value'].replace(",", ""))
st['duration'] = str(s['dur'])
for f in LockssCrawlStatus.INTFIELDMAP:
val = s[LockssCrawlStatus.INTFIELDMAP[f]]
if (val.__class__ == unicode):
st[f] = int(val.replace(",", ""))
else:
st[f] = str(val['value'].replace(",", ""))
except KeyError:
raise LockssError("CrawlStatusTable returned faulty info for %s: %s" % (auId.auId, s));
try:
# update existing crawl status
crawl = LockssCrawlStatus.objects.get( auId = auId, startTime = st['startTime'])
crawl.__dict__.update(st)
log.debug("LockssCrawlStatus UPD %s %s" % (crawl.startTime, str(crawl.auId)))
except ObjectDoesNotExist:
# create new crawlstatus
crawl = LockssCrawlStatus.objects.create(**st)
log.debug("LockssCrawlStatus NEW %s %s" % (crawl.startTime, str(crawl.auId)))
crawl.save()
