def fetchYumRepo(self, basepath="./", callback=None, verify_options=None):
startTime = time.time()
self.yumFetch = RepoFetch(self.repo_label, repourl=self.repo_url, \
cacert=self.sslcacert, clicert=self.sslclientcert, \
clikey=self.sslclientkey, mirrorlist=self.mirrors, \
download_dir=basepath, proxy_url=self.proxy_url, \
proxy_port=self.proxy_port, proxy_user=self.proxy_user, \
proxy_pass=self.proxy_pass, sslverify=self.sslverify,
max_speed=self.max_speed,
verify_options=verify_options)
self.fetchPkgs = ParallelFetch(self.yumFetch, self.numThreads, callback=callback)
try:
if not verify_options:
verify_options = {"size":False, "checksum":False}
self.yumFetch.setupRepo()
# first fetch the metadata
self.fetchPkgs.processCallback(ProgressReport.DownloadMetadata)
self.yumFetch.getRepoData()
if self.stopped:
return None
if not self.skip.has_key('packages') or self.skip['packages'] != 1:
# get rpms to fetch
self.prepareRPMS()
# get drpms to fetch
self.prepareDRPMS()
else:
log("Skipping packages preparation from sync process")
if not self.skip.has_key('distribution') or self.skip['distribution'] != 1:
# get Trees to fetch
self.prepareTrees()
else:
log("Skipping distribution preparation from sync process")
# prepare for download
self.fetchPkgs.addItemList(self.downloadinfo)
self.fetchPkgs.start()
report = self.fetchPkgs.waitForFinish()
self.yumFetch.finalizeMetadata()
endTime = time.time()
#log("Processed <%s> items in [%d] seconds" % (len(self.downloadinfo), \
# (endTime - startTime)))
if not self.skip.has_key('packages') or self.skip['packages'] != 1:
if self.purge_orphaned:
#log("Cleaning any orphaned packages..")
self.fetchPkgs.processCallback(ProgressReport.PurgeOrphanedPackages)
self.purgeOrphanPackages(self.yumFetch.getPackageList(), self.yumFetch.repo_dir)
if self.remove_old:
log("Removing old packages to limit to %s" % self.numOldPackages)
self.fetchPkgs.processCallback(ProgressReport.RemoveOldPackages)
gutils = GrinderUtils()
gutils.runRemoveOldPackages(self.pkgsavepath, self.numOldPackages)
self.yumFetch.deleteBaseCacheDir()
#log("Processed <%s> in %s seconds" % (report, endTime - startTime))
return report, (endTime - startTime)
finally:
self.fetchPkgs.stop()
self.yumFetch.closeRepo()
def setup(self, basepath="./", callback=None, verify_options=None, num_retries=None, retry_delay=None):
"""
Fetches yum metadata and determines what object should be downloaded.
@param basepath: path to store repo data
@type basepath: str
@param callback: progress callback function
@type callback: function which accepts a grinder.GrinderCallback.ProgressReport
@param verify_options: controls verification checks on "size" and "checksum".
@type verify_options: dict{"size":bool,"checksum":bool}
@param num_retries: number of retries to perform if an error occurs
@type num_retries: int
@param retry_delay: delay in seconds between retries, delay = 'retry_attempt' * 'retry_delay'
@type retry_delay: int
"""
self.repo_dir = os.path.join(basepath, self.repo_label)
LOG.info("%s, %s, Calling RepoFetch with: cacert=<%s>, clicert=<%s>, clikey=<%s>, proxy_url=<%s>, proxy_port=<%s>, proxy_user=<%s>, proxy_pass=<NOT_LOGGED>, sslverify=<%s>, max_speed=<%s>, verify_options=<%s>, filter=<%s>" %\
(self.repo_label, self.repo_url, self.sslcacert, self.sslclientcert, self.sslclientkey, self.proxy_url, self.proxy_port, self.proxy_user, self.sslverify, self.max_speed, verify_options, self.filter))
self.repoFetch = RepoFetch(cacert=self.sslcacert, clicert=self.sslclientcert, clikey=self.sslclientkey,\
proxy_url=self.proxy_url, proxy_port=self.proxy_port,
proxy_user=self.proxy_user, proxy_pass=self.proxy_pass,
sslverify=self.sslverify,
max_speed=self.max_speed,
verify_options=verify_options, num_retries=num_retries)
self.fetchPkgs = ParallelFetch(self.repoFetch, self.numThreads, callback=callback)
self.fetchPkgs.processCallback(ProgressReport.DownloadMetadata)
info = YumInfo(
repo_label=self.repo_label, repo_url=self.repo_url,
mirrors = self.mirrors, repo_dir=self.repo_dir,
packages_location=self.pkgpath, newest=self.newest,
remove_old=self.remove_old, numOldPackages=self.numOldPackages,
cacert=self.sslcacert, clicert=self.sslclientcert,
clikey=self.sslclientkey, proxy_url=self.proxy_url,
proxy_port=self.proxy_port, proxy_user=self.proxy_user,
proxy_pass=self.proxy_pass, sslverify=self.sslverify, skip=self.skip,
tmp_path=self.tmp_path, filter=self.filter,
num_retries=num_retries, retry_delay=retry_delay)
info.setUp()
self.rpmlist = info.rpms
self.drpmlist = info.drpms
def fetch(self, basepath="./", callback=None):
LOG.info("fetch basepath = %s" % (basepath))
startTime = time.time()
self.fileFetch = FileFetch(self.repo_label, self.repo_url, cacert=self.sslcacert, \
clicert=self.sslclientcert, clikey=self.sslclientkey, \
download_dir=basepath, proxy_url=self.proxy_url, \
proxy_port=self.proxy_port, proxy_user=self.proxy_user, \
proxy_pass=self.proxy_pass, sslverify=self.sslverify, max_speed=self.max_speed)
self.parallel_fetch_files = ParallelFetch(self.fileFetch,
self.numThreads,
callback=callback)
LOG.info("Determining downloadable Content bits...")
self.parallel_fetch_files.processCallback(
ProgressReport.DownloadMetadata)
self.prepareFiles()
# prepare for download
self.parallel_fetch_files.addItemList(self.downloadinfo)
self.parallel_fetch_files.start()
report = self.parallel_fetch_files.waitForFinish()
endTime = time.time()
LOG.info("Processed <%s> items in [%d] seconds" % (len(self.downloadinfo), \
(endTime - startTime)))
return report
def setup(self, basepath="./", callback=None, verify_options=None, num_retries=None, retry_delay=None, incr_progress=False):
"""
Fetches yum metadata and determines what object should be downloaded.
@param basepath: path to store repo data
@type basepath: str
@param callback: progress callback function
@type callback: function which accepts a grinder.GrinderCallback.ProgressReport
@param verify_options: controls verification checks on "size" and "checksum".
@type verify_options: dict{"size":bool,"checksum":bool}
@param num_retries: number of retries to perform if an error occurs
@type num_retries: int
@param retry_delay: delay in seconds between retries, delay = 'retry_attempt' * 'retry_delay'
@type retry_delay: int
@param incr_progress: if true, incremental progress on each item as it's downloaded will be reported
@type inc_progress: bool
"""
self.repo_dir = os.path.join(basepath, self.repo_label)
LOG.info("%s, %s, Calling RepoFetch with: cacert=<%s>, clicert=<%s>, clikey=<%s>, proxy_url=<%s>, proxy_port=<%s>, proxy_user=<%s>, proxy_pass=<NOT_LOGGED>, sslverify=<%s>, max_speed=<%s>, verify_options=<%s>, filter=<%s>" %\
(self.repo_label, self.repo_url, self.sslcacert, self.sslclientcert, self.sslclientkey, self.proxy_url, self.proxy_port, self.proxy_user, self.sslverify, self.max_speed, verify_options, self.filter))
self.repoFetch = RepoFetch(cacert=self.sslcacert, clicert=self.sslclientcert, clikey=self.sslclientkey,\
proxy_url=self.proxy_url, proxy_port=self.proxy_port,
proxy_user=self.proxy_user, proxy_pass=self.proxy_pass,
sslverify=self.sslverify,
max_speed=self.max_speed,
verify_options=verify_options, num_retries=num_retries)
self.fetchPkgs = ParallelFetch(self.repoFetch, self.numThreads, callback=callback, incr_progress=incr_progress)
self.fetchPkgs.processCallback(ProgressReport.DownloadMetadata)
info = YumInfo(
repo_label=self.repo_label, repo_url=self.repo_url,
mirrors = self.mirrors, repo_dir=self.repo_dir,
packages_location=self.pkgpath, newest=self.newest,
remove_old=self.remove_old, numOldPackages=self.numOldPackages,
cacert=self.sslcacert, clicert=self.sslclientcert,
clikey=self.sslclientkey, proxy_url=self.proxy_url,
proxy_port=self.proxy_port, proxy_user=self.proxy_user,
proxy_pass=self.proxy_pass, sslverify=self.sslverify, skip=self.skip,
tmp_path=self.tmp_path, filter=self.filter,
num_retries=num_retries, retry_delay=retry_delay)
info.setUp()
self.rpmlist = info.rpms
self.drpmlist = info.drpms
def syncPackages(self, channelLabel, savePath, verbose=0):
"""
channelLabel - channel to sync packages from
savePath - path to save packages
verbose - if true display more output
"""
startTime = time.time()
if channelLabel == "":
LOG.critical("No channel label specified to sync, abort sync.")
raise NoChannelLabelException()
LOG.info("sync(%s, %s) invoked" % (channelLabel, verbose))
satDump = SatDumpClient(self.baseURL, verbose=verbose)
LOG.debug("*** calling product_names ***")
packages = satDump.getChannelPackages(self.systemid, channelLabel)
LOG.info("%s packages are available, getting list of short metadata now." % (len(packages)))
pkgInfo = satDump.getShortPackageInfo(self.systemid, packages, filterLatest = not self.fetchAll)
LOG.info("%s packages have been marked to be fetched" % (len(pkgInfo.values())))
numThreads = int(self.parallel)
LOG.info("Running in parallel fetch mode with %s threads" % (numThreads))
pkgFetch = PackageFetch(self.systemid, self.baseURL, channelLabel, savePath)
self.parallelFetchPkgs = ParallelFetch(pkgFetch, numThreads)
self.parallelFetchPkgs.addItemList(pkgInfo.values())
self.parallelFetchPkgs.start()
report = self.parallelFetchPkgs.waitForFinish()
LOG.debug("Attempting to fetch comps.xml info from RHN")
self.fetchCompsXML(savePath, channelLabel)
self.fetchUpdateinfo(savePath, channelLabel)
endTime = time.time()
LOG.info("Processed <%s> %s packages, %s errors, completed in %s seconds" \
% (channelLabel, report.successes, report.errors, (endTime-startTime)))
if self.removeOldPackages:
LOG.info("Remove old packages from %s" % (savePath))
self.runRemoveOldPackages(savePath)
return report
def syncKickstarts(self, channelLabel, savePath, verbose=0):
"""
channelLabel - channel to sync kickstarts from
savePath - path to save kickstarts
verbose - if true display more output
"""
startTime = time.time()
satDump = SatDumpClient(self.baseURL, verbose=verbose)
ksLabels = satDump.getKickstartLabels(self.systemid, [channelLabel])
LOG.info("Found %s kickstart labels for channel %s" % (len(ksLabels[channelLabel]), channelLabel))
ksFiles = []
for ksLbl in ksLabels[channelLabel]:
LOG.info("Syncing kickstart label: %s" % (ksLbl))
metadata = satDump.getKickstartTreeMetadata(self.systemid, [ksLbl])
LOG.info("Retrieved metadata on %s files for kickstart label: %s" % (len(metadata[ksLbl]["files"]), ksLbl))
ksSavePath = os.path.join(savePath, ksLbl)
for ksFile in metadata[ksLbl]["files"]:
info = {}
info["relative-path"] = ksFile["relative-path"]
info["size"] = ksFile["file-size"]
info["md5sum"] = ksFile["md5sum"]
info["ksLabel"] = ksLbl
info["channelLabel"] = channelLabel
info["savePath"] = ksSavePath
ksFiles.append(info)
ksFetch = KickstartFetch(self.systemid, self.baseURL)
numThreads = int(self.parallel)
self.parallelFetchKickstarts = ParallelFetch(ksFetch, numThreads)
self.parallelFetchKickstarts.addItemList(ksFiles)
self.parallelFetchKickstarts.start()
report = self.parallelFetchKickstarts.waitForFinish()
endTime = time.time()
LOG.info("Processed %s %s %s kickstart files, %s errors, completed in %s seconds" \
% (channelLabel, ksLabels[channelLabel], report.successes,
report.errors, (endTime-startTime)))
return report
def fetch(self,basepath="./", callback=None):
LOG.info("fetch basepath = %s" % (basepath))
startTime = time.time()
self.fileFetch = FileFetch(self.repo_label, self.repo_url, cacert=self.sslcacert, \
clicert=self.sslclientcert, clikey=self.sslclientkey, \
download_dir=basepath, proxy_url=self.proxy_url, \
proxy_port=self.proxy_port, proxy_user=self.proxy_user, \
proxy_pass=self.proxy_pass, sslverify=self.sslverify, max_speed=self.max_speed)
self.parallel_fetch_files = ParallelFetch(self.fileFetch, self.numThreads, callback=callback)
LOG.info("Determining downloadable Content bits...")
self.parallel_fetch_files.processCallback(ProgressReport.DownloadMetadata)
self.prepareFiles()
# prepare for download
self.parallel_fetch_files.addItemList(self.downloadinfo)
self.parallel_fetch_files.start()
report = self.parallel_fetch_files.waitForFinish()
endTime = time.time()
LOG.info("Processed <%s> items in [%d] seconds" % (len(self.downloadinfo), \
(endTime - startTime)))
return report
class YumRepoGrinder(object):
"""
Driver class to fetch content from a Yum Repository
"""
def __init__(self, repo_label, repo_url, parallel=10, mirrors=None,
newest=False, cacert=None, clicert=None, clikey=None,
proxy_url=None, proxy_port=None, proxy_user=None,
proxy_pass=None, sslverify=1, packages_location=None,
remove_old=False, numOldPackages=2, skip=None, max_speed=None,
purge_orphaned=True, distro_location=None, tmp_path=None,
filter=None):
self.repo_label = repo_label
self.repo_url = repo_url
self.repo_dir = None
self.mirrors = mirrors
self.numThreads = int(parallel)
self.fetchPkgs = None
self.downloadinfo = []
self.repoFetch = None
self.fetchPkgs = None
self.sslcacert = cacert
self.sslclientcert = clicert
self.sslclientkey = clikey
self.temp_ssl_client_cert = None
self.temp_ssl_client_key = None
self.proxy_url = proxy_url
self.proxy_port = proxy_port
self.proxy_user = proxy_user
self.proxy_pass = proxy_pass
self.newest = newest
# set this if you want all packages to be stored in a central location
self.pkgpath = packages_location
self.numOldPackages = numOldPackages
self.pkgsavepath = ''
self.remove_old = remove_old
self.skip = skip
if not self.skip:
self.skip = []
self.sslverify = sslverify
self.max_speed = max_speed
self.purge_orphaned = purge_orphaned
self.stopped = False
self.distropath = distro_location
self.rpmlist = []
self.drpmlist = []
self.tmp_path = tmp_path
self.filter = filter
def getRPMItems(self):
return self.rpmlist
def getDeltaRPMItems(self):
return self.drpmlist
def getDistroItems(self):
return self.distro_items
def setup(self, basepath="./", callback=None, verify_options=None, num_retries=None, retry_delay=None, incr_progress=False):
"""
Fetches yum metadata and determines what object should be downloaded.
@param basepath: path to store repo data
@type basepath: str
@param callback: progress callback function
@type callback: function which accepts a grinder.GrinderCallback.ProgressReport
@param verify_options: controls verification checks on "size" and "checksum".
@type verify_options: dict{"size":bool,"checksum":bool}
@param num_retries: number of retries to perform if an error occurs
@type num_retries: int
@param retry_delay: delay in seconds between retries, delay = 'retry_attempt' * 'retry_delay'
@type retry_delay: int
@param incr_progress: if true, incremental progress on each item as it's downloaded will be reported
@type inc_progress: bool
"""
self.repo_dir = os.path.join(basepath, self.repo_label)
LOG.info("%s, %s, Calling RepoFetch with: cacert=<%s>, clicert=<%s>, clikey=<%s>, proxy_url=<%s>, proxy_port=<%s>, proxy_user=<%s>, proxy_pass=<NOT_LOGGED>, sslverify=<%s>, max_speed=<%s>, verify_options=<%s>, filter=<%s>" %\
(self.repo_label, self.repo_url, self.sslcacert, self.sslclientcert, self.sslclientkey, self.proxy_url, self.proxy_port, self.proxy_user, self.sslverify, self.max_speed, verify_options, self.filter))
self.repoFetch = RepoFetch(cacert=self.sslcacert, clicert=self.sslclientcert, clikey=self.sslclientkey,\
proxy_url=self.proxy_url, proxy_port=self.proxy_port,
proxy_user=self.proxy_user, proxy_pass=self.proxy_pass,
sslverify=self.sslverify,
max_speed=self.max_speed,
verify_options=verify_options, num_retries=num_retries)
self.fetchPkgs = ParallelFetch(self.repoFetch, self.numThreads, callback=callback, incr_progress=incr_progress)
self.fetchPkgs.processCallback(ProgressReport.DownloadMetadata)
info = YumInfo(
repo_label=self.repo_label, repo_url=self.repo_url,
mirrors = self.mirrors, repo_dir=self.repo_dir,
packages_location=self.pkgpath, newest=self.newest,
remove_old=self.remove_old, numOldPackages=self.numOldPackages,
cacert=self.sslcacert, clicert=self.sslclientcert,
clikey=self.sslclientkey, proxy_url=self.proxy_url,
proxy_port=self.proxy_port, proxy_user=self.proxy_user,
proxy_pass=self.proxy_pass, sslverify=self.sslverify, skip=self.skip,
tmp_path=self.tmp_path, filter=self.filter,
num_retries=num_retries, retry_delay=retry_delay)
info.setUp()
self.rpmlist = info.rpms
self.drpmlist = info.drpms
def setupDistroInfo(self):
info = DistroInfo(repo_url=self.repo_url, repo_dir=self.repo_dir,
distropath=self.distropath)
distro_items = info.prepareTrees(self.repoFetch)
self.distro_items = {}
if distro_items:
self.distro_items = distro_items
def addItems(self, items):
self.fetchPkgs.addItemList(items)
def download(self):
"""
Synchronous call, initiates download and waits for all items to finish before returning
@return: A SyncReport
@rtype: grinder.ParallelFetch.SyncReport
"""
try:
startTime = time.time()
self.fetchPkgs.start()
self.fetchPkgs.processCallback(ProgressReport.DownloadItems)
report = self.fetchPkgs.waitForFinish()
self.finalizeMetadata()
if 'rpm' not in self.skip:
if self.purge_orphaned:
# Includes logic of:
# 1) removing previously existing packages that have been
# removed from repository metadata
# 2) removing old packages that are part of repository metadata,
# but we want removed because of remove_old/numOldPackages option
LOG.info("Cleaning any orphaned packages..")
self.fetchPkgs.processCallback(ProgressReport.PurgeOrphanedPackages)
self.purgeOrphanPackages()
if self.remove_old:
# Need to re-test remove_old is functioning
# We added the ability to limit the old packages from being downloaded
# I think we need to address the case of existing packages from a prior sync
self.fetchPkgs.processCallback(ProgressReport.RemoveOldPackages)
endTime = time.time()
LOG.info("Processed <%s>,<%s> with <%s> items in [%d] seconds. Report: %s" % (self.repo_label, self.repo_url, len(self.downloadinfo),\
(endTime - startTime), report))
return report
finally:
if self.fetchPkgs:
self.fetchPkgs.stop()
self.fetchPkgs = None
def fetchYumRepo(self, basepath="./", callback=None, verify_options=None):
LOG.info("fetchYumRepo() repo_label = %s, repo_url = %s, basepath = %s, verify_options = %s" % \
(self.repo_label, self.repo_url, basepath, verify_options))
self.setup(basepath, callback, verify_options)
if 'distribution' not in self.skip:
self.setupDistroInfo()
if self.distro_items:
self.addItems(self.distro_items['files'])
else:
LOG.debug("skipping distributions from sync")
self.addItems(self.rpmlist)
self.addItems(self.drpmlist)
return self.download()
def stop(self, block=True):
LOG.info("Stopping")
self.stopped = True
if self.fetchPkgs:
self.fetchPkgs.stop()
if block:
LOG.info("Block is <%s> so waiting" % (block))
self.fetchPkgs._waitForThreads()
def purgeOrphanPackages(self):
"""
While re-sync purge any orphaned packages in the Repo that we did not intend to sync.
Includes removal of packages no longer in primary.xml
as well as older packages filtered out from remove_old/numOldPackages logic
"""
dpkgs = []
if self.rpmlist:
for pkg in self.rpmlist:
dpkgs.append(os.path.join(self.repo_dir, os.path.dirname(pkg['relativepath']), pkg['fileName']))
if os.path.exists(self.repo_dir):
for root, dirs, files in os.walk(self.repo_dir):
for f in files:
tmp_path = os.path.join(root, f)
if tmp_path.endswith('.rpm') and tmp_path not in dpkgs:
LOG.info("Removing orphan package: %s" % (tmp_path))
os.remove(tmp_path)
def finalizeMetadata(self):
local_repo_path = "%s/%s" % (self.repo_dir, "repodata")
local_new_path = "%s/%s" % (self.repo_dir, "repodata.new")
if not os.path.exists(local_new_path):
LOG.info("No new metadata to finalize.")
return
try:
LOG.info("Finalizing metadata, moving %s to %s" % (local_new_path, local_repo_path))
if os.path.exists(local_repo_path):
# remove existing metadata before copying
shutil.rmtree(local_repo_path)
shutil.copytree(local_new_path, local_repo_path)
shutil.rmtree(local_new_path)
except Exception, e:
LOG.error("An error occurred while finalizing metadata:\n%s" % str(e))
class YumRepoGrinder(object):
"""
Driver class to fetch content from a Yum Repository
"""
def __init__(self, repo_label, repo_url, parallel=10, mirrors=None,
newest=False, cacert=None, clicert=None, clikey=None,
proxy_url=None, proxy_port=None, proxy_user=None,
proxy_pass=None, sslverify=1, packages_location=None,
remove_old=False, numOldPackages=2, skip=None, max_speed=None,
purge_orphaned=True, distro_location=None, tmp_path=None,
filter=None):
self.repo_label = repo_label
self.repo_url = repo_url
self.repo_dir = None
self.mirrors = mirrors
self.numThreads = int(parallel)
self.fetchPkgs = None
self.downloadinfo = []
self.repoFetch = None
self.fetchPkgs = None
self.sslcacert = cacert
self.sslclientcert = clicert
self.sslclientkey = clikey
self.temp_ssl_client_cert = None
self.temp_ssl_client_key = None
self.proxy_url = proxy_url
self.proxy_port = proxy_port
self.proxy_user = proxy_user
self.proxy_pass = proxy_pass
self.newest = newest
# set this if you want all packages to be stored in a central location
self.pkgpath = packages_location
self.numOldPackages = numOldPackages
self.pkgsavepath = ''
self.remove_old = remove_old
self.skip = skip
if not self.skip:
self.skip = []
self.sslverify = sslverify
self.max_speed = max_speed
self.purge_orphaned = purge_orphaned
self.stopped = False
self.distropath = distro_location
self.rpmlist = []
self.drpmlist = []
self.tmp_path = tmp_path
self.filter = filter
def getRPMItems(self):
return self.rpmlist
def getDeltaRPMItems(self):
return self.drpmlist
def getDistroItems(self):
return self.distro_items
def setup(self, basepath="./", callback=None, verify_options=None, num_retries=None, retry_delay=None):
"""
Fetches yum metadata and determines what object should be downloaded.
@param basepath: path to store repo data
@type basepath: str
@param callback: progress callback function
@type callback: function which accepts a grinder.GrinderCallback.ProgressReport
@param verify_options: controls verification checks on "size" and "checksum".
@type verify_options: dict{"size":bool,"checksum":bool}
@param num_retries: number of retries to perform if an error occurs
@type num_retries: int
@param retry_delay: delay in seconds between retries, delay = 'retry_attempt' * 'retry_delay'
@type retry_delay: int
"""
self.repo_dir = os.path.join(basepath, self.repo_label)
LOG.info("%s, %s, Calling RepoFetch with: cacert=<%s>, clicert=<%s>, clikey=<%s>, proxy_url=<%s>, proxy_port=<%s>, proxy_user=<%s>, proxy_pass=<NOT_LOGGED>, sslverify=<%s>, max_speed=<%s>, verify_options=<%s>, filter=<%s>" %\
(self.repo_label, self.repo_url, self.sslcacert, self.sslclientcert, self.sslclientkey, self.proxy_url, self.proxy_port, self.proxy_user, self.sslverify, self.max_speed, verify_options, self.filter))
self.repoFetch = RepoFetch(cacert=self.sslcacert, clicert=self.sslclientcert, clikey=self.sslclientkey,\
proxy_url=self.proxy_url, proxy_port=self.proxy_port,
proxy_user=self.proxy_user, proxy_pass=self.proxy_pass,
sslverify=self.sslverify,
max_speed=self.max_speed,
verify_options=verify_options, num_retries=num_retries)
self.fetchPkgs = ParallelFetch(self.repoFetch, self.numThreads, callback=callback)
self.fetchPkgs.processCallback(ProgressReport.DownloadMetadata)
info = YumInfo(
repo_label=self.repo_label, repo_url=self.repo_url,
mirrors = self.mirrors, repo_dir=self.repo_dir,
packages_location=self.pkgpath, newest=self.newest,
remove_old=self.remove_old, numOldPackages=self.numOldPackages,
cacert=self.sslcacert, clicert=self.sslclientcert,
clikey=self.sslclientkey, proxy_url=self.proxy_url,
proxy_port=self.proxy_port, proxy_user=self.proxy_user,
proxy_pass=self.proxy_pass, sslverify=self.sslverify, skip=self.skip,
tmp_path=self.tmp_path, filter=self.filter,
num_retries=num_retries, retry_delay=retry_delay)
info.setUp()
self.rpmlist = info.rpms
self.drpmlist = info.drpms
def setupDistroInfo(self):
info = DistroInfo(repo_url=self.repo_url, repo_dir=self.repo_dir,
distropath=self.distropath)
distro_items = info.prepareTrees(self.repoFetch)
self.distro_items = {}
if distro_items:
self.distro_items = distro_items
def addItems(self, items):
self.fetchPkgs.addItemList(items)
def download(self):
"""
Synchronous call, initiates download and waits for all items to finish before returning
@return: A SyncReport
@rtype: grinder.ParallelFetch.SyncReport
"""
try:
startTime = time.time()
self.fetchPkgs.start()
self.fetchPkgs.processCallback(ProgressReport.DownloadItems)
report = self.fetchPkgs.waitForFinish()
self.finalizeMetadata()
if 'rpm' not in self.skip:
if self.purge_orphaned:
# Includes logic of:
# 1) removing previously existing packages that have been
# removed from repository metadata
# 2) removing old packages that are part of repository metadata,
# but we want removed because of remove_old/numOldPackages option
LOG.info("Cleaning any orphaned packages..")
self.fetchPkgs.processCallback(ProgressReport.PurgeOrphanedPackages)
self.purgeOrphanPackages()
if self.remove_old:
# Need to re-test remove_old is functioning
# We added the ability to limit the old packages from being downloaded
# I think we need to address the case of existing packages from a prior sync
self.fetchPkgs.processCallback(ProgressReport.RemoveOldPackages)
endTime = time.time()
LOG.info("Processed <%s>,<%s> with <%s> items in [%d] seconds. Report: %s" % (self.repo_label, self.repo_url, len(self.downloadinfo),\
(endTime - startTime), report))
return report
finally:
if self.fetchPkgs:
self.fetchPkgs.stop()
self.fetchPkgs = None
def fetchYumRepo(self, basepath="./", callback=None, verify_options=None):
LOG.info("fetchYumRepo() repo_label = %s, repo_url = %s, basepath = %s, verify_options = %s" % \
(self.repo_label, self.repo_url, basepath, verify_options))
self.setup(basepath, callback, verify_options)
if 'distribution' not in self.skip:
self.setupDistroInfo()
if self.distro_items:
self.addItems(self.distro_items['files'])
else:
LOG.debug("skipping distributions from sync")
self.addItems(self.rpmlist)
self.addItems(self.drpmlist)
return self.download()
def stop(self, block=True):
LOG.info("Stopping")
self.stopped = True
if self.fetchPkgs:
self.fetchPkgs.stop()
if block:
LOG.info("Block is <%s> so waiting" % (block))
self.fetchPkgs._waitForThreads()
def purgeOrphanPackages(self):
"""
While re-sync purge any orphaned packages in the Repo that we did not intend to sync.
Includes removal of packages no longer in primary.xml
as well as older packages filtered out from remove_old/numOldPackages logic
"""
dpkgs = []
if self.rpmlist:
for pkg in self.rpmlist:
dpkgs.append(os.path.join(self.repo_dir, os.path.dirname(pkg['relativepath']), pkg['fileName']))
if os.path.exists(self.repo_dir):
for root, dirs, files in os.walk(self.repo_dir):
for f in files:
tmp_path = os.path.join(root, f)
if tmp_path.endswith('.rpm') and tmp_path not in dpkgs:
LOG.info("Removing orphan package: %s" % (tmp_path))
os.remove(tmp_path)
def finalizeMetadata(self):
local_repo_path = "%s/%s" % (self.repo_dir, "repodata")
local_new_path = "%s/%s" % (self.repo_dir, "repodata.new")
if not os.path.exists(local_new_path):
LOG.info("No new metadata to finalize.")
return
try:
LOG.info("Finalizing metadata, moving %s to %s" % (local_new_path, local_repo_path))
if os.path.exists(local_repo_path):
# remove existing metadata before copying
shutil.rmtree(local_repo_path)
shutil.copytree(local_new_path, local_repo_path)
shutil.rmtree(local_new_path)
except Exception, e:
LOG.error("An error occurred while finalizing metadata:\n%s" % str(e))
class FileGrinder(object):
"""
Driver module to initiate the file fetching
"""
def __init__(self, repo_label, url, parallel=50, cacert=None, clicert=None, clikey=None, \
proxy_url=None, proxy_port=None, proxy_user=None, \
proxy_pass=None, sslverify=1, files_location=None, max_speed=None ):
self.repo_label = repo_label
self.repo_url = url
self.numThreads = int(parallel)
self.downloadinfo = []
self.sslcacert = cacert
self.sslclientcert = clicert
self.sslclientkey = clikey
self.proxy_url = proxy_url
self.proxy_port = proxy_port
self.proxy_user = proxy_user
self.proxy_pass = proxy_pass
# set this if you want all packages to be stored in a central location
self.filepath = files_location
self.sslverify = sslverify
self.max_speed = max_speed
self.fileFetch = None
def prepareFiles(self):
file_manifest = "PULP_MANIFEST"
file_url = self.fileFetch.url + '/' + file_manifest
file_name = file_manifest
file_path = self.fileFetch.repo_dir
info = {
'downloadurl' : file_url,
'fileName' : file_name,
'savepath' : file_path,
'checksumtype' : None,
'checksum' : None,
'size' : None,
'pkgpath' : None,
}
self.fileFetch.fetchItem(info)
file_info = {}
file_manifest_path = os.path.join(file_path, file_manifest)
if os.path.exists(file_manifest_path):
file_info = parseManifest(file_manifest_path)
else:
LOG.info("File Metadata Not Found at url %s" % self.repo_url)
for fileinfo in file_info:
info = {}
info['downloadurl'] = self.repo_url + '/' + fileinfo['filename']
info['fileName'] = os.path.basename(fileinfo['filename'])
info['savepath'] = file_path #+ '/' + os.path.dirname(info['filename'])
info['checksumtype'] = 'sha256'
info['checksum'] = fileinfo['checksum']
info['size'] = int(fileinfo['size'])
if self.filepath:
info['pkgpath'] = "%s/%s/%s/%s/" % (self.filepath, os.path.basename(fileinfo['filename'])[:3], \
os.path.basename(fileinfo['filename']), fileinfo['checksum'])
else:
info['pkgpath'] = None
info['item_type'] = BaseFetch.FILE
self.downloadinfo.append(info)
LOG.info("%s files have been marked to be fetched" % len(file_info))
def fetch(self,basepath="./", callback=None):
LOG.info("fetch basepath = %s" % (basepath))
startTime = time.time()
self.fileFetch = FileFetch(self.repo_label, self.repo_url, cacert=self.sslcacert, \
clicert=self.sslclientcert, clikey=self.sslclientkey, \
download_dir=basepath, proxy_url=self.proxy_url, \
proxy_port=self.proxy_port, proxy_user=self.proxy_user, \
proxy_pass=self.proxy_pass, sslverify=self.sslverify, max_speed=self.max_speed)
self.parallel_fetch_files = ParallelFetch(self.fileFetch, self.numThreads, callback=callback)
LOG.info("Determining downloadable Content bits...")
self.parallel_fetch_files.processCallback(ProgressReport.DownloadMetadata)
self.prepareFiles()
# prepare for download
self.parallel_fetch_files.addItemList(self.downloadinfo)
self.parallel_fetch_files.start()
report = self.parallel_fetch_files.waitForFinish()
endTime = time.time()
LOG.info("Processed <%s> items in [%d] seconds" % (len(self.downloadinfo), \
(endTime - startTime)))
return report
def stop(self, block=True):
if self.parallel_fetch_files:
self.parallel_fetch_files.stop()
if block:
self.parallel_fetch_files._waitForThreads()
class FileGrinder(object):
"""
Driver module to initiate the file fetching
"""
def __init__(self, repo_label, url, parallel=50, cacert=None, clicert=None, clikey=None, \
proxy_url=None, proxy_port=None, proxy_user=None, \
proxy_pass=None, sslverify=1, files_location=None, max_speed=None ):
self.repo_label = repo_label
self.repo_url = url
self.numThreads = int(parallel)
self.downloadinfo = []
self.sslcacert = cacert
self.sslclientcert = clicert
self.sslclientkey = clikey
self.proxy_url = proxy_url
self.proxy_port = proxy_port
self.proxy_user = proxy_user
self.proxy_pass = proxy_pass
# set this if you want all packages to be stored in a central location
self.filepath = files_location
self.sslverify = sslverify
self.max_speed = max_speed
self.fileFetch = None
def prepareFiles(self):
file_manifest = "PULP_MANIFEST"
file_url = self.fileFetch.url + '/' + file_manifest
file_name = file_manifest
file_path = self.fileFetch.repo_dir
info = {
'downloadurl': file_url,
'fileName': file_name,
'savepath': file_path,
'checksumtype': None,
'checksum': None,
'size': None,
'pkgpath': None,
}
self.fileFetch.fetchItem(info)
file_info = {}
file_manifest_path = os.path.join(file_path, file_manifest)
if os.path.exists(file_manifest_path):
file_info = parseManifest(file_manifest_path)
else:
LOG.info("File Metadata Not Found at url %s" % self.repo_url)
for fileinfo in file_info:
info = {}
info['downloadurl'] = self.repo_url + '/' + fileinfo['filename']
info['fileName'] = os.path.basename(fileinfo['filename'])
info[
'savepath'] = file_path #+ '/' + os.path.dirname(info['filename'])
info['checksumtype'] = 'sha256'
info['checksum'] = fileinfo['checksum']
info['size'] = int(fileinfo['size'])
if self.filepath:
info['pkgpath'] = "%s/%s/%s/%s/" % (self.filepath, os.path.basename(fileinfo['filename'])[:3], \
os.path.basename(fileinfo['filename']), fileinfo['checksum'])
else:
info['pkgpath'] = None
info['item_type'] = BaseFetch.FILE
self.downloadinfo.append(info)
LOG.info("%s files have been marked to be fetched" % len(file_info))
def fetch(self, basepath="./", callback=None):
LOG.info("fetch basepath = %s" % (basepath))
startTime = time.time()
self.fileFetch = FileFetch(self.repo_label, self.repo_url, cacert=self.sslcacert, \
clicert=self.sslclientcert, clikey=self.sslclientkey, \
download_dir=basepath, proxy_url=self.proxy_url, \
proxy_port=self.proxy_port, proxy_user=self.proxy_user, \
proxy_pass=self.proxy_pass, sslverify=self.sslverify, max_speed=self.max_speed)
self.parallel_fetch_files = ParallelFetch(self.fileFetch,
self.numThreads,
callback=callback)
LOG.info("Determining downloadable Content bits...")
self.parallel_fetch_files.processCallback(
ProgressReport.DownloadMetadata)
self.prepareFiles()
# prepare for download
self.parallel_fetch_files.addItemList(self.downloadinfo)
self.parallel_fetch_files.start()
report = self.parallel_fetch_files.waitForFinish()
endTime = time.time()
LOG.info("Processed <%s> items in [%d] seconds" % (len(self.downloadinfo), \
(endTime - startTime)))
return report
def stop(self, block=True):
if self.parallel_fetch_files:
self.parallel_fetch_files.stop()
if block:
self.parallel_fetch_files._waitForThreads()
class Grinder:
def __init__(self, url, username, password, cert, systemid, parallel, verbose):
self.baseURL = url
self.cert = open(cert, 'r').read()
self.systemid = open(systemid, 'r').read()
self.username = username
self.password = password
self.parallel = parallel
self.fetchAll = False #default is only fetch latest packages
self.parallelFetchPkgs = None
self.parallelFetchKickstarts = None
self.skipProductList = []
self.skipPackageList = []
self.verbose = verbose
self.killcount = 0
self.removeOldPackages = False
self.numOldPkgsKeep = 1
self.rhnComm = RHNComm(url, self.systemid)
def setRemoveOldPackages(self, value):
self.removeOldPackages = value
def getRemoveOldPackages(self):
return self.removeOldPackages
def getFetchAllPackages(self):
return self.fetchAll
def setFetchAllPackages(self, val):
self.fetchAll = val
def getSkipProductList(self):
return self.skipProductList
def setSkipProductList(self, skipProductList):
self.skipProductList = skipProductList
def getSkipPackageList(self):
return self.skipPackageList
def setSkipPackageList(self, skipPackageList):
self.skipPackageList = skipPackageList
def getNumOldPackagesToKeep(self):
return self.numOldPkgsKeep
def setNumOldPackagesToKeep(self, num):
self.numOldPkgsKeep = num
def deactivate(self):
SATELLITE_URL = "%s/rpc/api" % (self.baseURL)
client = getRhnApi(SATELLITE_URL, verbose=0)
key = client.auth.login(self.username, self.password)
retval = client.satellite.deactivateSatellite(self.systemid)
print "retval from deactivation: %s" % retval
client.auth.logout(key)
print "Deactivated!"
def activate(self):
rhn = RHNTransport()
satClient = getRhnApi(self.baseURL + "/SAT",
verbose=self.verbose, transport=rhn)
# First check if we are active
active = False
retval = satClient.authentication.check(self.systemid)
LOG.debug("AUTH CHECK: %s " % str(retval))
if (retval == 1):
LOG.debug("We are activated ... continue!")
active = True
else:
LOG.debug("Not active")
if (not active):
if(not self.username or not self.password):
raise SystemNotActivatedException()
SATELLITE_URL = "%s/rpc/api" % (self.baseURL)
client = RhnApi(SATELLITE_URL, verbose=0)
key = client.auth.login(self.username, self.password)
retval = client.satellite.activateSatellite(self.systemid, self.cert)
LOG.debug("retval from activation: %s" % retval)
if (retval != 1):
raise CantActivateException()
client.auth.logout(key)
LOG.debug("Activated!")
def stop(self):
if (self.parallelFetchPkgs):
self.parallelFetchPkgs.stop()
if (self.parallelFetchKickstarts):
self.parallelFetchKickstarts.stop()
def checkChannels(self, channelsToSync):
"""
Input:
channelsToSync - list of channels to sync
Output:
list containing bad channel names
"""
satDump = SatDumpClient(self.baseURL)
channelFamilies = satDump.getChannelFamilies(self.systemid)
badChannel = []
for channelLabel in channelsToSync:
found = False
for d in channelFamilies.values():
if channelLabel in d["channel_labels"]:
LOG.debug("Found %s under %s" % (channelLabel, d["label"]))
found = True
break
if not found:
LOG.debug("Unable to find %s, adding it to badChannel list" % (channelLabel))
badChannel.append(channelLabel)
return badChannel
def getChannelLabels(self):
labels = {}
satDump = SatDumpClient(self.baseURL)
channelFamilies = satDump.getChannelFamilies(self.systemid)
for d in channelFamilies.values():
if (d["label"] in self.skipProductList):
continue
labels[d["label"]] = d["channel_labels"]
return labels
def displayListOfChannels(self):
labels = self.getChannelLabels()
print("List of channels:")
for lbl in labels:
print("\nProduct : %s\n" % (lbl))
for l in labels[lbl]:
print(" %s" % (l))
def syncKickstarts(self, channelLabel, savePath, verbose=0):
"""
channelLabel - channel to sync kickstarts from
savePath - path to save kickstarts
verbose - if true display more output
"""
startTime = time.time()
satDump = SatDumpClient(self.baseURL, verbose=verbose)
ksLabels = satDump.getKickstartLabels(self.systemid, [channelLabel])
LOG.info("Found %s kickstart labels for channel %s" % (len(ksLabels[channelLabel]), channelLabel))
ksFiles = []
for ksLbl in ksLabels[channelLabel]:
LOG.info("Syncing kickstart label: %s" % (ksLbl))
metadata = satDump.getKickstartTreeMetadata(self.systemid, [ksLbl])
LOG.info("Retrieved metadata on %s files for kickstart label: %s" % (len(metadata[ksLbl]["files"]), ksLbl))
ksSavePath = os.path.join(savePath, ksLbl)
for ksFile in metadata[ksLbl]["files"]:
info = {}
info["relative-path"] = ksFile["relative-path"]
info["size"] = ksFile["file-size"]
info["md5sum"] = ksFile["md5sum"]
info["ksLabel"] = ksLbl
info["channelLabel"] = channelLabel
info["savePath"] = ksSavePath
ksFiles.append(info)
ksFetch = KickstartFetch(self.systemid, self.baseURL)
numThreads = int(self.parallel)
self.parallelFetchKickstarts = ParallelFetch(ksFetch, numThreads)
self.parallelFetchKickstarts.addItemList(ksFiles)
self.parallelFetchKickstarts.start()
report = self.parallelFetchKickstarts.waitForFinish()
endTime = time.time()
LOG.info("Processed %s %s %s kickstart files, %s errors, completed in %s seconds" \
% (channelLabel, ksLabels[channelLabel], report.successes,
report.errors, (endTime-startTime)))
return report
def syncPackages(self, channelLabel, savePath, verbose=0):
"""
channelLabel - channel to sync packages from
savePath - path to save packages
verbose - if true display more output
"""
startTime = time.time()
if channelLabel == "":
LOG.critical("No channel label specified to sync, abort sync.")
raise NoChannelLabelException()
LOG.info("sync(%s, %s) invoked" % (channelLabel, verbose))
satDump = SatDumpClient(self.baseURL, verbose=verbose)
LOG.debug("*** calling product_names ***")
packages = satDump.getChannelPackages(self.systemid, channelLabel)
LOG.info("%s packages are available, getting list of short metadata now." % (len(packages)))
pkgInfo = satDump.getShortPackageInfo(self.systemid, packages, filterLatest = not self.fetchAll)
LOG.info("%s packages have been marked to be fetched" % (len(pkgInfo.values())))
numThreads = int(self.parallel)
LOG.info("Running in parallel fetch mode with %s threads" % (numThreads))
pkgFetch = PackageFetch(self.systemid, self.baseURL, channelLabel, savePath)
self.parallelFetchPkgs = ParallelFetch(pkgFetch, numThreads)
self.parallelFetchPkgs.addItemList(pkgInfo.values())
self.parallelFetchPkgs.start()
report = self.parallelFetchPkgs.waitForFinish()
LOG.debug("Attempting to fetch comps.xml info from RHN")
self.fetchCompsXML(savePath, channelLabel)
self.fetchUpdateinfo(savePath, channelLabel)
endTime = time.time()
LOG.info("Processed <%s> %s packages, %s errors, completed in %s seconds" \
% (channelLabel, report.successes, report.errors, (endTime-startTime)))
if self.removeOldPackages:
LOG.info("Remove old packages from %s" % (savePath))
self.runRemoveOldPackages(savePath)
return report
def fetchCompsXML(self, savePath, channelLabel):
###
# Fetch comps.xml, used by createrepo for "groups" info
###
compsxml = ""
try:
compsxml = self.rhnComm.getRepodata(channelLabel, "comps.xml")
except GetRequestException, ge:
if (ge.code == 404):
LOG.info("Channel has no compsXml")
else:
raise ge
if not savePath:
savePath = channelLabel
f = open(os.path.join(savePath, "comps.xml"), "w")
f.write(compsxml)
f.close()
