def load_keystrings(keystrings):
for auid in keystrings:
if (not MasterAuId.objects.filter(auId=auid).exists()):
try:
MasterAuId(auId=auid).save()
except LockssError as err:
log.warn("could not create: %s" % (err))
def load(cache, auId, doUrls, expire, trials, sleep, timeout):
act = Action.GETAUSUMMARY
if (doUrls):
act = act + "," + Action.GETURLLIST
log.info('Start %s: %s expire=%s ...' % (act, auId, str(expire)))
success = False
try:
for _ in range(trials):
try:
if (doUrls):
UrlReport.loadTry(cache.ui, auId, expire)
else:
LockssCacheAuSummary.__loadTry(cache.ui, auId, expire)
success = True
break
except urllib2.HTTPError as inst:
cache.reconnect(sleep,timeout)
except ExpatError:
log.error("XML Parser error; could not %s %s" % (auId, act))
success = False; # try again
if (not success):
log.error("exhausted trials for %s; could not load %s" % (auId, act))
except LockssError as inst:
log.warn("LockssException: %s" % inst) # output is scanned for the ERROR string
log.warn("could not digest %s for %s" % (act, auId.auId))
finally:
log.debug2('Stop %s: %s Success = %s ...' % (act, auId, success))
return success
def printConfigFile(fname, action, auIds, cache):
log.warn("printing ids for %s on %s to %s" % (action, cache, fname))
f = open(fname, 'w')
f.write('[%s]\n' % LockssScript.PARAMSECTION)
f.write('actionlist = %s\n' % action)
f.write('auidlist =\n')
for au in auIds:
f.write('\t' + au.auId + "\n ")
f.write('\n')
f.close()
def load(cache, auId, trials, sleep,timeout):
success = False
log.debug2('Start %s: %s ...' % (Action.GETCRAWLSTATUS, auId))
try:
log.info('get %s: %s ...' % (Action.GETCRAWLSTATUS, auId))
for i in range(trials):
try:
LockssCrawlStatus.__loadTry(cache.ui, auId)
success = True
break
except urllib2.HTTPError as inst:
cache.reconnect(sleep,timeout)
log.error("exhausted trials for %s, could not load crawlstatus" % (auId))
except LockssError as inst:
log.warn("LockssException: %s" % inst) # output is scanned for the ERROR string
log.warn("could not digest %s for %s" % (Action.GETCRAWLSTATUS, auId.auId))
finally:
log.debug2('Stop %s: %s Success = %s ...' % (Action.GETCRAWLSTATUS, auId, success))
return success
def printcsv(folder, auids, orderby, hdrs, sep, minrev = 1):
'''
print url reports for all given auids including urls that have a
version at least as great as minrev, which defaults to 1
'''
if (not auids):
log.info('NOOP %s: No auids to print to %s' % (Action.PRTURLLIST, folder) )
return
for auid in auids:
urls = []
try:
if (orderby == 'minversion' or orderby == 'replication'):
urls = auid.urlreport.url_set.filter(version__gte=minrev).all()
else:
urls = auid.urlreport.url_set.filter(version__gte=minrev).order_by(orderby).all()
ext = ".tsv"
if (sep == ","):
ext = ".csv"
f = open(folder + "/" + auid.auId + ext, 'w')
if (urls.count() == 0):
log.info("NOOP %s: file %s No Urls with version >= %s" % (Action.PRTURLLIST, f.name, minrev))
log.info('Start %s: file %s version %s' % (Action.PRTURLLIST, f.name, minrev))
try:
reportDate = auid.urlreport.reportDate
f.write("ReportDate\t%s\nIncluding Urls with version >= %s\n\n" % (str(reportDate), minrev))
f.write(sep.join(hdrs) + "\n")
for url in urls:
f.write(url.csv(hdrs, sep) + "\n")
log.debug2('Stop %s: file %s version %s' % (Action.PRTURLLIST, f.name, minrev))
f.close()
except IndexError:
log.info("NOOP %s: file %s No Urls at all" % (Action.PRTURLLIST, f.name))
except ObjectDoesNotExist:
log.warn('Start %s: No UrlReport for %s at %s' %
(Action.PRTURLLIST, auid, auid.cache.name))
def process(self):
log.info("---")
log.info("Start Processing")
if (self.options.dryrun):
return
opts = self.options.__dict__
print "# COMMAND", self.options._COMMAND;
for key in ['ausetname', 'auseturl']:
if (opts.has_key(key)):
print "#", key, opts.get(key);
print "# ";
caches = self.get_caches();
print "# SERVER\t", "\n# SERVER\t".join([str(c) for c in caches]);
print "# ";
# TODO deal with serverlist - aka restrict to aus on given servers
fields = ["auid", "repl", "reportDate", "cache", "agree",
"sizeMB", "diskMB", "repository",
"#urls", "avg_version", "min_version", "max_version",
"best_copy"]
print "\t".join(fields);
for auid in self.options.auids:
# TODO restrict to servers in options.serverlist
auids = LockssCacheAuId.objects.filter(Q(auId=auid) ,self.get_Q_caches() )
log.info("AUID: %s" % "\t".join([auid, "#matches=%d" % auids.count()]));
profiles = []
for au in auids:
prof = {
'auid' : auid,
'repl' : au.replication(),
'reportDate' : "",
'cache' : au.cache.name,
'agree' : "",
'repository' : "",
'#urls' : "",
'avg_version' : "",
'min_version' : "",
'max_version' : "",
'sizeMB': "",
'diskMB': "",
"best_copy" : False
}
lausum = LockssCacheAuSummary.objects.filter(auId = au);
if (not lausum.exists()):
log.warn("No AuSummary Info for " + str(au))
else:
lausum = lausum[0];
if (lausum.agreement != None) : prof['agree'] = lausum.agreement
prof['repository'] = lausum.repository
prof['sizeMB'] = lausum.contentSizeMB()
prof['diskMB'] = lausum.diskUsageMB
prof['reportDate'] = lausum.reportDate
urlReport = UrlReport.objects.filter(auId = au);
if (not urlReport.exists()):
log.warn("No Url Info for " + str(au))
prof['nurls'] = 0
else:
report = urlReport[0];
urls = report.url_set;
version_info = urls.aggregate(Avg('version'), Min('version'), Max('version'))
prof['avg_version'] = version_info['version__avg'];
prof['min_version'] = version_info['version__min'];
prof['max_version'] = version_info['version__max'];
profiles.append(prof)
# find the au that has the max agreement and if there are multiple
# the one/one of the ones with the max avg_version number
# and designate as best_copy
#
# find max avg_version number
if (not [] == profiles):
max_agree = max(v['agree'] for v in profiles)
# find all that have that max_agree value
candidates = []
for prof in profiles:
if (prof['agree'] == max_agree):
candidates.append(prof)
max_version = max(v['avg_version'] for v in candidates)
for candidate in profiles:
if (candidate['avg_version'] == max_version):
candidate['best_copy'] = 'True'
break
assert(candidate)
for prof in profiles:
vals = [];
for f in fields:
v = prof[f]
if (v.__class__ == float):
vals.append("%.2f" % v)
else:
vals.append(str(v))
print "\t".join(vals)
for i in range(0,3):
print "";
log.info("Stop Processing")
