Commit Graph

12639 Commits

Author SHA1 Message Date
Nick Mathewson
50b06a2b76 make the description of tolen_asserts more dire
We have a CVE # for this bug.
2011-01-15 11:49:26 -05:00
Nick Mathewson
115782bdbe Fix a heap overflow found by debuger, and make it harder to make that mistake again
Our public key functions assumed that they were always writing into a
large enough buffer.  In one case, they weren't.

(Incorporates fixes from sebastian)
2011-01-15 11:49:25 -05:00
Nick Mathewson
a16902b9d4 Always nul-terminate the result passed to evdns_server_add_ptr_reply
In dnsserv_resolved(), we carefully made a nul-terminated copy of the
answer in a PTR RESOLVED cell... then never used that nul-terminated
copy.  Ouch.

Surprisingly this one isn't as huge a security problem as it could be.
The only place where the input to dnsserv_resolved wasn't necessarily
nul-terminated was when it was called indirectly from relay.c with the
contents of a relay cell's payload.  If the end of the payload was
filled with junk, eventdns.c would take the strdup() of the name [This
part is bad; we might crash there if the cell is in a bad part of the
stack or the heap] and get a name of at least length
495[*]. eventdns.c then rejects any name of length over 255, so the
bogus data would be neither transmitted nor altered.

  [*] If the name was less than 495 bytes long, the client wouldn't
     actually be reading off the end of the cell.

Nonetheless this is a reasonably annoying bug.  Better fix it.

Found while looking at bug 2332, reported by doorss.  Bugfix on
0.2.0.1-alpha.
2011-01-15 11:49:25 -05:00
Nick Mathewson
1f3b442023 catch another overlong malloc possibility. found by cypherpunks 2011-01-15 10:42:11 -05:00
Nick Mathewson
b27f5cc50d Fix another instance of "128" in buffers.c. More bug2330. 2011-01-15 10:25:58 -05:00
Nick Mathewson
2c04c506a4 Merge remote branch 'origin/maint-0.2.1' into maint-0.2.2 2011-01-12 14:38:11 -05:00
Nick Mathewson
9fcc14224b Make our replacement INT32_MAX always signed
The C standard says that INT32_MAX is supposed to be a signed
integer.  On platforms that have it, we get the correct
platform-defined value.  Our own replacement, however, was
unsigned.  That's going to cause a bug somewhere eventually.
2011-01-12 14:29:38 -05:00
Nick Mathewson
43e12a7988 Merge remote branch 'public/bug2363' into maint-0.2.2 2011-01-12 14:00:20 -05:00
Nick Mathewson
a2c41aa3e8 Merge remote branch 'sebastian/bug2337' into maint-0.2.2 2011-01-12 12:55:09 -05:00
Nick Mathewson
1e37d8678a Merge branch 'bug2331' into maint-0.2.2 2011-01-12 12:42:40 -05:00
Nick Mathewson
f4756051fe add a missing "not" in bug2331 changelog 2011-01-12 12:42:21 -05:00
Nick Mathewson
597433bcec Merge branch 'bug2346' into maint-0.2.2 2011-01-12 12:37:51 -05:00
Nick Mathewson
5044cb9752 Explain bug2346 fix better based on suggestions from arma 2011-01-12 12:37:42 -05:00
Sebastian Hahn
2dd7df8308 Fix a autoconf warning 2011-01-12 04:02:52 +01:00
Nick Mathewson
aa45e82593 Pull up more data when parsing socks messages
Previously, we only looked at up to 128 bytes.  This is a bad idea
since socks messages can be at least 256+x bytes long.  Now we look at
up to 512 bytes; this should be enough for 0.2.2.x to handle all valid
SOCKS messages.  For 0.2.3.x, we can think about handling trickier
cases.

Fixes 2330.  Bugfix on 0.2.0.16-alpha.
2011-01-10 17:24:16 -05:00
Nick Mathewson
95968a625e Wait 60 minutes before retrying failed state save; bug2346 2011-01-10 16:51:11 -05:00
Nick Mathewson
bd67b23fec Always nul-terminate the result passed to evdns_server_add_ptr_reply
In dnsserv_resolved(), we carefully made a nul-terminated copy of the
answer in a PTR RESOLVED cell... then never used that nul-terminated
copy.  Ouch.

Surprisingly this one isn't as huge a security problem as it could be.
The only place where the input to dnsserv_resolved wasn't necessarily
nul-terminated was when it was called indirectly from relay.c with the
contents of a relay cell's payload.  If the end of the payload was
filled with junk, eventdns.c would take the strdup() of the name [This
part is bad; we might crash there if the cell is in a bad part of the
stack or the heap] and get a name of at least length
495[*]. eventdns.c then rejects any name of length over 255, so the
bogus data would be neither transmitted nor altered.

  [*] If the name was less than 495 bytes long, the client wouldn't
     actually be reading off the end of the cell.

Nonetheless this is a reasonably annoying bug.  Better fix it.

Found while looking at bug 2332, reported by doorss.  Bugfix on
0.2.0.1-alpha.
2011-01-10 16:33:48 -05:00
Nick Mathewson
80253ba4d7 fix some spelling in rephist comments 2011-01-10 15:13:07 -05:00
Nick Mathewson
105b94b75b Add Maxima lists to bandwidth state.
Right now, Tor routers don't save the maxima values from the
bw_history_t between sessions.  That's no good, since we use those
values to determine bandwidth.  This code adds a new BWHist.*Maximum
set of values to the state file.  If they're not present, we estimate
them by taking the observed total bandwidth and dividing it by the
period length, which provides a lower bound.

This should fix bug 1863.  I'm calling it a feature.
2011-01-10 14:47:39 -05:00
Nick Mathewson
5ae391762c Merge remote branch 'origin/maint-0.2.1' into maint-0.2.2 2011-01-10 14:12:33 -05:00
Nick Mathewson
28844c8403 On bwhist state load failure, clear dir_read/write hist too 2011-01-10 13:15:51 -05:00
Nick Mathewson
7e1502c0d1 Fix bug when parsing bwhist with unexpected Interval
Previously, our state parsing code would fail to parse a bwhist
correctly if the Interval was anything other than the default
hardcoded 15 minutes.  This change makes the parsing less incorrect,
though the resulting history array might get strange values in it if
the intervals don't match the one we're using.  (That is, if stuff was
generated in 15 minute intervals, and we read it into an array that
expects 30 minute intervals, we're fine, since values can be combined
pairwise.  But if we generate data at 30 minute intervals and read it
into 15 minute intervals, alternating buckets will be empty.)

Bugfix on 0.1.1.11-alpha.
2011-01-10 13:06:50 -05:00
Nick Mathewson
8dd4ecd14e Turn the loop bodies of rep_hist_save/load_state into functions
The trick of looping from i=0..4 , switching on i to set up some
variables, then running some common code is much better expressed by
just calling a function 4 times with 4 sets of arguments.  This should
make the code a little easier to follow and maintain here.
2011-01-10 12:59:07 -05:00
Nick Mathewson
373a1bc40e Impose maximum sizes on parsed objects
An object, you'll recall, is something between -----BEGIN----- and
-----END----- tags in a directory document.  Some of our code, as
doorss has noted in bug 2352, could assert if one of these ever
overflowed SIZE_T_CEILING but not INT_MAX.  As a solution, I'm setting
a maximum size on a single object such that neither of these limits
will ever be hit.  I'm also fixing the INT_MAX checks, just to be sure.
2011-01-10 12:12:11 -05:00
Nick Mathewson
729f404efe Add logic in routerparse to not read overlong private keys
I am not at all sure that it is possible to trigger a bug here,
but better safe than sorry.
2011-01-10 12:07:34 -05:00
Karsten Loesing
d43cba6c69 Update to the January 1 2011 Maxmind GeoLite Country database. 2011-01-10 10:37:54 +01:00
Nick Mathewson
efc9a84108 Add missing parens to evdns_base_resolve_* macros while I am at it 2011-01-09 19:05:06 -05:00
Nick Mathewson
68d5b6bc52 Fix check for failed evdns request creation
When using libevent 2, we use evdns_base_resolve_*().  When not, we
fake evdns_base_resolve_*() using evdns_resolve_*().

Our old check was looking for negative values (like libevent 2
returns), but our eventdns.c code returns 1.  This code makes the
check just test for nonzero.

Note that this broken check was not for _resolve_ failures or even for
failures to _launch_ a resolve: it was for failures to _create_ or
_encode_ a resolve request.

Bug introduced in 81eee0ecfff3dac1e9438719d2f7dc0ba7e84a71; found by
lodger; uploaded to trac by rransom.  Bug 2363.  Fix on 0.2.2.6-alpha.
2011-01-09 19:02:57 -05:00
Nick Mathewson
732275bb81 Describe tor-resolve defaults. Bug 2364. 2011-01-09 15:40:40 -05:00
Nick Mathewson
54135b72f8 Merge remote branch 'origin/maint-0.2.1' into maint-0.2.2 2011-01-07 22:04:40 -05:00
Nick Mathewson
045e6ebd31 Remove a loud info log message 2011-01-07 22:03:22 -05:00
Nick Mathewson
0a35ac6a22 Correctly detect and exclude addresses outside of our virtual address range
Found by cypherpunks; fixes more of 2328.  Bug was introduced in 3623a122;
first appeared in 0.2.0.5-alpha.
2011-01-07 12:24:36 -05:00
Nick Mathewson
d6329eda96 Merge remote branch 'origin/maint-0.2.1' into maint-0.2.2 2011-01-06 13:37:39 -05:00
Nick Mathewson
d6b49c55c5 Merge branch 'bug2328_021' into maint-0.2.1 2011-01-06 13:36:29 -05:00
Nick Mathewson
2008728df7 Notice a little faster if we're running out of virtual addresses
We were not decrementing "available" every time we did
++next_virtual_addr in addressmap_get_virtual_address: we left out the
--available when we skipped .00 and .255 addresses.

This didn't actually cause a bug in most cases, since the failure mode
was to keep looping around the virtual addresses until we found one,
or until available hit zero.  It could have given you an infinite loop
rather than a useful message, however, if you said "VirtualAddrNetwork
127.0.0.255/32" or something broken like that.

Spotted by cypherpunks
2011-01-06 13:29:36 -05:00
Nick Mathewson
eabddd8ca0 Handle a NULL return from addressmap_get_virtual_address
Fix for bug 2328; bugfix on 0.1.2.1-alpha; bug found by doorss.
2011-01-05 16:36:48 -05:00
Nick Mathewson
31d6659d97 Fix a double-counting bug in addrmap_get_virtual_address
We were decrementing "available" twice for each in-use address we ran
across.  This would make us declare that we ran out of virtual
addresses when the address space was only half full.
2011-01-05 16:02:43 -05:00
Nick Mathewson
240fa42aac Fix size_t vs unsigned comparison too 2011-01-05 12:49:02 -05:00
Nick Mathewson
d14b0d54d2 Fix a SIZE_T_CEILING check in torgzip.c; noticed by cypherpunks 2011-01-05 12:42:34 -05:00
Nick Mathewson
0e9d7f105d Merge remote branch 'origin/maint-0.2.1' into maint-0.2.2 2011-01-03 17:24:32 -05:00
Nick Mathewson
e4320689cc Note that Tor requires Automake 1.7. Fixes bug 2305 2011-01-03 17:24:16 -05:00
Nick Mathewson
81d69f4c2d Detect signed size_t and report an error at configure time. 2011-01-03 16:54:57 -05:00
Nick Mathewson
0222228d64 Fix up size and sign issues in base32 code
Fixes bug 2331.
2011-01-03 16:16:53 -05:00
Nick Mathewson
64798dab4f Detect and disallow compression bombs 2011-01-03 15:54:23 -05:00
Nick Mathewson
f089804332 Merge remote branch 'origin/maint-0.2.1' into maint-0.2.2 2011-01-03 15:31:19 -05:00
Nick Mathewson
e365aee971 Avoid assertion on read_file_to_str() with size==SIZE_T_CEILING-1
Spotted by doors, fixes bug 2326.
2011-01-03 15:30:11 -05:00
Nick Mathewson
cee433d751 Merge remote branch 'origin/maint-0.2.1' into maint-0.2.2 2011-01-03 15:15:54 -05:00
Nick Mathewson
e09ab69703 Check size against SIZE_T_CEILING in realloc too.
Fixes bug 2324.
2011-01-03 15:15:27 -05:00
Nick Mathewson
27cefef3a2 Merge remote branch 'origin/maint-0.2.1' into maint-0.2.2 2011-01-03 13:18:33 -05:00
Nick Mathewson
bb5f99d4df Merge remote branch 'sebastian/bug2314' into maint-0.2.2 2011-01-03 12:47:14 -05:00