Some distributors ship CD and DVD files with SHA1 hash sums instead
of MD5 hash sums, so let's extend the kvm_utils functions to
evaluate and compare SHA1 hashes:
* hash_file(): Calculate a hash sum for a file, supports SHA1 check as well,
replaces md5sum_file()
* unmap_url_cache(): Reimplementation of a function present on
autotest utils that downloads a file and caches it. The reason
I'm keeping it is that I want more testing before I move all
needed function definitions to the autotest API
Also, extend VM.create() method to support SHA1 check.
2nd try, addressing Uri's comments.
Signed-off-by: Lucas Meneghel Rodrigues <lmr@xxxxxxxxxx>
---
client/tests/kvm/calc_md5sum_1m.py | 2 +-
client/tests/kvm/kvm_utils.py | 105 ++++++++++++++++++++++++++++++++----
client/tests/kvm/kvm_vm.py | 20 +++++---
3 files changed, 108 insertions(+), 19 deletions(-)
diff --git a/client/tests/kvm/calc_md5sum_1m.py b/client/tests/kvm/calc_md5sum_1m.py
index 2325673..153a1e0 100755
--- a/client/tests/kvm/calc_md5sum_1m.py
+++ b/client/tests/kvm/calc_md5sum_1m.py
@@ -18,4 +18,4 @@ else:
if not os.access(fname, os.F_OK) or not os.access(fname, os.R_OK):
print 'bad file name or permissions'
else:
- print kvm_utils.md5sum_file(fname, 1024*1024)
+ print kvm_utils.hash_file(fname, 1024*1024, method="md5")
diff --git a/client/tests/kvm/kvm_utils.py b/client/tests/kvm/kvm_utils.py
index 53b664a..f72984a 100644
--- a/client/tests/kvm/kvm_utils.py
+++ b/client/tests/kvm/kvm_utils.py
@@ -4,8 +4,8 @@ KVM test utility functions.
@copyright: 2008-2009 Red Hat Inc.
"""
-import md5, thread, subprocess, time, string, random, socket, os, signal, pty
-import select, re, logging, commands, cPickle
+import md5, sha, thread, subprocess, time, string, random, socket, os, signal
+import select, re, logging, commands, cPickle, pty
from autotest_lib.client.bin import utils
from autotest_lib.client.common_lib import error
import kvm_subprocess
@@ -760,23 +760,35 @@ def wait_for(func, timeout, first=0.0, step=1.0, text=None):
return None
-def md5sum_file(filename, size=None):
+def hash_file(filename, size=None, method="md5"):
"""
- Calculate the md5sum of filename.
+ Calculate the hash of filename.
If size is not None, limit to first size bytes.
Throw exception if something is wrong with filename.
Can be also implemented with bash one-liner (assuming size%1024==0):
- dd if=filename bs=1024 count=size/1024 | md5sum -
+ dd if=filename bs=1024 count=size/1024 | sha1sum -
- @param filename: Path of the file that will have its md5sum calculated.
- @param returns: md5sum of the file.
+ @param filename: Path of the file that will have its hash calculated.
+ @param method: Method used to calculate the hash. Supported methods:
+ * md5
+ * sha1
+ @returns: Hash of the file, if something goes wrong, return None.
"""
chunksize = 4096
fsize = os.path.getsize(filename)
- if not size or size>fsize:
+
+ if not size or size > fsize:
size = fsize
f = open(filename, 'rb')
- o = md5.new()
+
+ if method == "md5":
+ hash = md5.new()
+ elif method == "sha1":
+ hash = sha.new()
+ else:
+ logging.error("Unknown hash type %s, returning None" % method)
+ return None
+
while size > 0:
if chunksize > size:
chunksize = size
@@ -784,7 +796,78 @@ def md5sum_file(filename, size=None):
if len(data) == 0:
logging.debug("Nothing left to read but size=%d" % size)
break
- o.update(data)
+ hash.update(data)
size -= len(data)
f.close()
- return o.hexdigest()
+ return hash.hexdigest()
+
+
+def get_hash_from_file(hash_path, dvd_basename):
+ """
+ Get the a hash from a given DVD image from a hash file
+ (Hash files are usually named MD5SUM or SHA1SUM and are located inside the
+ download directories of the DVDs)
+
+ @param hash_path: Local path to a hash file.
+ @param cd_image: Basename of a CD image
+ """
+ hash_file = open(hash_path, 'r')
+ for line in hash_file.readlines():
+ if dvd_basename in line:
+ return line.split()[0]
+
+
+def unmap_url_cache(cachedir, url, expected_hash, method="md5"):
+ """
+ Downloads a file from a URL to a cache directory. If the file is already
+ at the expected position and has the expected hash, let's not download it
+ again.
+
+ @param cachedir: Directory that might hold a copy of the file we want to
+ download.
+ @param url: URL for the file we want to download.
+ @param expected_hash: Hash string that we expect the file downloaded to
+ have.
+ @param method: Method used to calculate the hash string (md5, sha1).
+ """
+ # Let's convert cachedir to a canonical path, if it's not already
+ cachedir = os.path.realpath(cachedir)
+ if not os.path.isdir(cachedir):
+ try:
+ os.makedirs(cachedir)
+ except:
+ raise ValueError('Could not create cache directory %s' % cachedir)
+ file_from_url = os.path.basename(url)
+ file_local_path = os.path.join(cachedir, file_from_url)
+
+ file_hash = None
+ failure_counter = 0
+ while not file_hash == expected_hash:
+ if os.path.isfile(file_local_path):
+ if method == "md5":
+ file_hash = hash_file(file_local_path, method="md5")
+ elif method == "sha1":
+ file_hash = hash_file(file_local_path, method="sha1")
+
+ if file_hash == expected_hash:
+ # File is already at the expected position and ready to go
+ src = file_from_url
+ else:
+ # Let's download the package again, it's corrupted...
+ logging.error("Seems that file %s is corrupted, trying to "
+ "download it again" % file_from_url)
+ src = url
+ failure_counter += 1
+ else:
+ # File is not there, let's download it
+ src = url
+ if failure_counter > 1:
+ raise EnvironmentError("Consistently failed to download the "
+ "package %s. Aborting further download "
+ "attempts. This might mean either the "
+ "network connection has problems or the "
+ "expected hash string that was determined "
+ "for this file is wrong" % file_from_url)
+ file_path = utils.unmap_url(cachedir, src, cachedir)
+
+ return file_path
diff --git a/client/tests/kvm/kvm_vm.py b/client/tests/kvm/kvm_vm.py
index a8d96ca..9aada61 100755
--- a/client/tests/kvm/kvm_vm.py
+++ b/client/tests/kvm/kvm_vm.py
@@ -339,20 +339,26 @@ class VM:
if params.get("md5sum_1m"):
logging.debug("Comparing expected MD5 sum with MD5 sum of "
"first MB of ISO file...")
- actual_md5sum = kvm_utils.md5sum_file(iso, 1048576)
- expected_md5sum = params.get("md5sum_1m")
+ actual_hash = kvm_utils.hash_file(iso, 1048576, method="md5")
+ expected_hash = params.get("md5sum_1m")
compare = True
elif params.get("md5sum"):
logging.debug("Comparing expected MD5 sum with MD5 sum of ISO "
"file...")
- actual_md5sum = kvm_utils.md5sum_file(iso)
- expected_md5sum = params.get("md5sum")
+ actual_hash = kvm_utils.md5sum_file(iso, method="md5")
+ expected_hash = params.get("md5sum")
+ compare = True
+ elif params.get("sha1sum"):
+ logging.debug("Comparing expected SHA1 sum with SHA1 sum of "
+ "ISO file...")
+ actual_hash = kvm_utils.md5sum_file(iso, method="sha1")
+ expected_hash = params.get("md5sum")
compare = True
if compare:
- if actual_md5sum == expected_md5sum:
- logging.debug("MD5 sums match")
+ if actual_hash == expected_hash:
+ logging.debug("Hashes match")
else:
- logging.error("Actual MD5 sum differs from expected one")
+ logging.error("Actual hash differs from expected one")
return False
# Make sure the following code is not executed by more than one thread
--
1.6.2.5
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
