Commit Graph

63 Commits

Author SHA1 Message Date
relyea%netscape.com
78033ccb01 Add the ability to test the multi-access database feature with existing test cases. 2002-04-08 23:39:44 +00:00
sonja.mirtitsch%sun.com
08ce0e0ee9 patch for bug 87914 r=wtc, NSS test scripts require . in the PATH 2002-03-20 03:01:27 +00:00
sonja.mirtitsch%sun.com
95d835e50a changing the way to figure out if selfserv is still present from ps to
kill -0 so we might be able to debug the disapearing selfserv problem
better
2002-02-06 05:34:16 +00:00
sonja.mirtitsch%sun.com
0f9e24e63e calling the selfserv now with -y to get the certs at the end, adjusted clients and wait times 2002-02-06 02:51:59 +00:00
wtc%netscape.com
eeef12be2f Bugzilla bug 119340: backed out the fix. We are still seeing the
"selfserv process not detectable" errors on Linux and I don't have
time to get to the bottom of it.
Modified files:
	cmd/selfserv/selfserv.c tests/common/init.sh tests/ssl/ssl.sh
2002-01-31 02:03:57 +00:00
wtc%netscape.com
bd2bc25ec1 Bugzilla bug 119340: an inelegant but more reliable way to kill the
multithreaded 'selfserv' process on Linux.
Modified files:
    cmd/selfserv/selfserv.c tests/common/init.sh tests/ssl/ssl.sh
2002-01-23 03:18:57 +00:00
sonja.mirtitsch%sun.com
8243a281b2 to avoid client and server writes at the same time write output to a tmp
file first, and then the tmp file to stdout
the writes had lead to garbled messages that were then interpreted as
errormessages by the scripts
2002-01-04 21:31:39 +00:00
sonja.mirtitsch%sun.com
97127f127f fix for enhancemet request 114964 2002-01-03 23:58:23 +00:00
sonja.mirtitsch%sun.com
8bc8d0113f fix for bug 73098, need to make other checkins in ssl.sh and don't
want to merge
Ian reviewed the patch
2001-12-14 20:02:53 +00:00
sonja.mirtitsch%sun.com
9c37d51af8 enabling the tls stresstest 2001-12-12 03:15:01 +00:00
sonja.mirtitsch%sun.com
4a8b2ae277 modernized... changed servers, clients, mastebuild, 10 minutes wait now 2001-11-15 23:27:20 +00:00
sonja.mirtitsch%sun.com
0c0eae99c2 next try to gat all of the HP problems... 2001-11-02 23:47:47 +00:00
sonja.mirtitsch%sun.com
b44b405186 put a workaround in place, so client ands server will not write at the
same time
(to avoid lines like: ...Unable to find the certificate
authenticatfailedion...)
2001-11-02 19:09:11 +00:00
sonja.mirtitsch%sun.com
9631ae5009 putting new tests back, since the ciphers have been fixed again 2001-09-27 16:37:46 +00:00
sonja.mirtitsch%sun.com
a3944b022e backing out changes until coredump of testclnt is fixed 2001-09-25 20:14:20 +00:00
sonja.mirtitsch%sun.com
3152497a31 added tests of 3 new ciphers in TLS and SSL 2001-09-25 16:00:00 +00:00
sonmi%netscape.com
3572f7c67c took out some outdated 3.2 workarounds
put in debug info for sjsu
2001-09-13 01:03:07 +00:00
sonmi%netscape.com
57589cceed fixed typo 2001-08-24 01:11:58 +00:00
sonmi%netscape.com
6b7463a799 took hardcoded reference to clio out 2001-08-20 21:25:56 +00:00
wtc%netscape.com
f2143ac49e Bugzilla bug 86528: have tstclnt exit with 254 (= -2 mod 256) where it
was exiting with -2 before so that old test script that expects 254
continues to work.  r=larryh.
Modified Files: tstclnt.c sslauth.txt
2001-06-27 00:29:01 +00:00
wtc%netscape.com
7481a570da Bugzilla bug #86528: fixed tstclnt so that it always exits with a
nonnegative status.  This way we can correctly extract its exit status on
Windows, so the temporary workaround in ssl.sh can be removed.  Changed
sslauth.txt to match the new exit status of tstclnt.  r=larryh.
Modified files:
   cmd/tstclnt/tstclnt.c tests/ssl/ssl.sh tests/ssl/sslauth.txt
2001-06-25 19:53:16 +00:00
sonmi%netscape.com
63fa350f6d removed -v option, made client and server write at the same time 2001-06-13 17:42:29 +00:00
sonmi%netscape.com
0cda219441 turned on verbose to see where (if) selfserver is dying 2001-06-12 23:29:53 +00:00
sonmi%netscape.com
227f854124 seperated the output of the client and the server to work around
a problem on dump
2001-05-16 00:08:05 +00:00
sonmi%netscape.com
4fcffdb12c yesterday's fix broke the backwardcompatibility tests because it used a new
option - put check for backwardcompatibility tests before
2001-05-11 23:05:32 +00:00
sonmi%netscape.com
adc43e9eae added -q to the stressclient, so the test will terminate with an error if
the selfsrv process disappears
2001-05-11 00:02:06 +00:00
sonmi%netscape.com
0b64a39959 added -D to the selfserver 2001-04-20 20:50:24 +00:00
sonmi%netscape.com
eeaea4c91d fixed a stresstest failures were not recognized by this script 2001-04-10 18:39:14 +00:00
sonmi%netscape.com
4fd12678a2 improved usage and parameters, reduced connections from 5 to 1 2001-04-02 17:32:48 +00:00
sonmi%netscape.com
3e4ef196fb changes for the new distributed stress test
-) to get around the DNS/ipaddress problem
   -) be sourcable from ssl_dist_stress.sh (share a lot of functionality)
2001-03-28 00:01:43 +00:00
sonmi%netscape.com
f9ebd65d76 distributed stress test
new script to run the server and the remote side of the test
   -) have cert.sh generate certs
   -) start a server
   -) launch strsclnts via rsh, each rsh with a range of certs to use
   -) wait 10 minutes
   -) shut down the server
2001-03-28 00:00:48 +00:00
sonmi%netscape.com
bed15077e8 fixed a typo in an echo 2001-03-03 01:59:29 +00:00
sonmi%netscape.com
1ae750085c complete restructuring and cleanup of the QA suites.
bug 65000 and 68878
2001-02-28 22:12:47 +00:00
sonmi%netscape.com
74241c3ef8 added information in the result.html when selfserver dies (previously
only in output.log)
2001-01-23 01:47:32 +00:00
sonmi%netscape.com
14b74a767a formated everything to a <tab4> format - still needs reformating to linelength 2001-01-11 01:55:23 +00:00
sonmi%netscape.com
672d2f9598 terminate now if no selfserver is running before attempting to start the
clients (with exception of the -q testclien)
2001-01-11 01:45:12 +00:00
sonmi%netscape.com
9b58d20702 removed certutilout with the other temporary files 2000-12-21 20:11:38 +00:00
sonmi%netscape.com
a63c3e7c14 removed the "fix" for a nonproblem :
the problem appeared to be a failure of the shell to pass on the correct return
codes when run from cron (false positive and false negative results when run
from cron - in reality the returncodes were right, but the cron generated db
had no password because of a bug in certutil
2000-12-20 23:40:53 +00:00
sonmi%netscape.com
b9c1650063 undid the redirecting to a tmp file, only write the stderr to the stdout 2000-12-20 00:24:38 +00:00
sonmi%netscape.com
eaeef3c20b explicitly redirected certutil output once more to get rid of cron error msg 2000-12-19 00:26:21 +00:00
sonmi%netscape.com
06ecbd3c25 fixed a comment, sed is very picky about strings starting with -n (it thinks
this is an option)
2000-12-15 01:07:45 +00:00
sonmi%netscape.com
2e5bb06d28 fixed unreasonable long PATH (which refused) and got rid of the error when
run interactively (-n_TestUser-w_bogus was filtered through an sed, which
thought that -n was **his** option...)
2000-12-12 05:10:51 +00:00
sonmi%netscape.com
9bf885bd61 added timestamps for stresstest per nelson's request 2000-12-08 23:49:55 +00:00
sonmi%netscape.com
f558ce20b2 took out the NULL_SERVER restart code completely 2000-11-29 01:09:33 +00:00
sonmi%netscape.com
8a8c1d8900 removed another server restart (NULL ciphers), but left code in in case we
need to reactivate the restart (search for variable NULL_SERVER)
2000-11-28 02:57:46 +00:00
sonmi%netscape.com
30455c891c added debug echo for each testclient started 2000-11-17 22:12:26 +00:00
sonmi%netscape.com
e40084da9d checked in previous fix again. Hanging had other reasons 2000-11-17 02:17:18 +00:00
sonmi%netscape.com
6365388287 backed out my changes, first it wanted input, then I started it <request file,
which made the next testclient fail / hang ...
2000-11-16 02:35:11 +00:00
sonmi%netscape.com
b40960af31 added -d CLIENTDIR to the new testclient starts, so NT won't have a problem 2000-11-15 22:20:53 +00:00
sonmi%netscape.com
ce41b46375 fixed a blank that that had gotten in the middle of a variablename 2000-11-15 00:48:36 +00:00