From: Brian Warner Date: Tue, 12 Feb 2008 01:17:01 +0000 (-0700) Subject: add 'tahoe catalog-shares' tool, to make a one-line summary of each share file. This... X-Git-Tag: allmydata-tahoe-0.8.0~95 X-Git-Url: https://git.rkrishnan.org/pf/content/install.html?a=commitdiff_plain;h=ca27b8e5ad311a5850140dc0f00352668419d467;p=tahoe-lafs%2Ftahoe-lafs.git add 'tahoe catalog-shares' tool, to make a one-line summary of each share file. This can help do cross-server correlation of sharefiles, looking for anomalies --- diff --git a/src/allmydata/scripts/debug.py b/src/allmydata/scripts/debug.py index 71060b08..f3da8576 100644 --- a/src/allmydata/scripts/debug.py +++ b/src/allmydata/scripts/debug.py @@ -339,15 +339,124 @@ def find_shares(config, out=sys.stdout, err=sys.stderr): return 0 + +class CatalogSharesOptions(usage.Options): + """ + Run this as 'catalog-shares NODEDIRS..', and it will emit a line to stdout + for each share it finds: + + CHK $SI $k/$N $filesize $UEB_hash $abspath_sharefile + SDMF $SI $k/$N $seqnum/$roothash $abspath_sharefile + UNKNOWN $abspath_sharefile + + It may be useful to build up a catalog of shares from many storage servers + and then sort the results. If you see shares with the same SI but different + parameters/filesize/UEB_hash, then something is wrong. + + """ + def parseArgs(self, *nodedirs): + self.nodedirs = nodedirs + +def describe_share(abs_sharefile, si_s, shnum_s, out, err): + from allmydata import uri, storage, mutable + from allmydata.util import idlib + import struct + + f = open(abs_sharefile, "rb") + prefix = f.read(32) + + if prefix == storage.MutableShareFile.MAGIC: + # mutable share + m = storage.MutableShareFile(abs_sharefile) + WE, nodeid = m._read_write_enabler_and_nodeid(f) + num_extra_leases = m._read_num_extra_leases(f) + data_length = m._read_data_length(f) + extra_lease_offset = m._read_extra_lease_offset(f) + container_size = extra_lease_offset - m.DATA_OFFSET + leases = list(m._enumerate_leases(f)) + + share_type = "unknown" + f.seek(m.DATA_OFFSET) + if f.read(1) == "\x00": + # this slot contains an SMDF share + share_type = "SDMF" + + if share_type == "SDMF": + f.seek(m.DATA_OFFSET) + data = f.read(min(data_length, 2000)) + + try: + pieces = mutable.unpack_share(data) + except mutable.NeedMoreDataError, e: + # retry once with the larger size + size = e.needed_bytes + f.seek(m.DATA_OFFSET) + data = f.read(min(data_length, size)) + pieces = mutable.unpack_share(data) + (seqnum, root_hash, IV, k, N, segsize, datalen, + pubkey, signature, share_hash_chain, block_hash_tree, + share_data, enc_privkey) = pieces + + print >>out, "SDMF %s %d/%d #%d:%s %s" % (si_s, k, N, seqnum, + idlib.b2a(root_hash), + abs_sharefile) + else: + print >>out, "UNKNOWN mutable %s" % (abs_sharefile,) + + elif struct.unpack(">L", prefix[:4]) == (1,): + # immutable + + sf = storage.ShareFile(abs_sharefile) + # use a ReadBucketProxy to parse the bucket and find the uri extension + bp = storage.ReadBucketProxy(None) + offsets = bp._parse_offsets(sf.read_share_data(0, 0x24)) + seek = offsets['uri_extension'] + length = struct.unpack(">L", sf.read_share_data(seek, 4))[0] + seek += 4 + UEB_data = sf.read_share_data(seek, length) + + unpacked = uri.unpack_extension_readable(UEB_data) + k = unpacked["needed_shares"] + N = unpacked["total_shares"] + filesize = unpacked["size"] + ueb_hash = unpacked["UEB_hash"] + + print >>out, "CHK %s %d/%d %d %s %s" % (si_s, k, N, filesize, ueb_hash, + abs_sharefile) + + else: + print >>out, "UNKNOWN really-unknown %s" % (abs_sharefile,) + + f.close() + + +def catalog_shares(config, out=sys.stdout, err=sys.stderr): + for d in config.nodedirs: + d = os.path.join(os.path.expanduser(d), "storage/shares") + if os.path.exists(d): + for abbrevdir in os.listdir(d): + abbrevdir = os.path.join(d, abbrevdir) + for si_s in os.listdir(abbrevdir): + si_dir = os.path.join(abbrevdir, si_s) + for shnum_s in os.listdir(si_dir): + abs_sharefile = os.path.join(si_dir, shnum_s) + assert os.path.isfile(abs_sharefile) + describe_share(abs_sharefile, si_s, shnum_s, out, err) + return 0 + + + subCommands = [ ["dump-share", None, DumpOptions, "Unpack and display the contents of a share (uri_extension and leases)."], ["dump-cap", None, DumpCapOptions, "Unpack a read-cap or write-cap"], ["find-shares", None, FindSharesOptions, "Locate sharefiles in node dirs"], + ["catalog-shares", None, CatalogSharesOptions, "Describe shares in node dirs"], ] dispatch = { "dump-share": dump_share, "dump-cap": dump_cap, "find-shares": find_shares, + "catalog-shares": catalog_shares, } diff --git a/src/allmydata/test/test_system.py b/src/allmydata/test/test_system.py index 71233b87..51d0f608 100644 --- a/src/allmydata/test/test_system.py +++ b/src/allmydata/test/test_system.py @@ -1190,6 +1190,20 @@ class SystemTest(testutil.SignalMixin, testutil.PollMixin, unittest.TestCase): sharefiles = [sfn.strip() for sfn in out.readlines()] self.failUnlessEqual(len(sharefiles), 10) + # also exercise the 'catalog-shares' tool + out,err = StringIO(), StringIO() + nodedirs = [self.getdir("client%d" % i) for i in range(self.numclients)] + cmd = ["catalog-shares"] + nodedirs + rc = runner.runner(cmd, stdout=out, stderr=err) + self.failUnlessEqual(rc, 0) + out.seek(0) + descriptions = [sfn.strip() for sfn in out.readlines()] + self.failUnlessEqual(len(descriptions), 30) + matching = [line + for line in descriptions + if line.startswith("CHK %s " % storage_index_s)] + self.failUnlessEqual(len(matching), 10) + def _test_control(self, res): # exercise the remote-control-the-client foolscap interfaces in # allmydata.control (mostly used for performance tests)