summaryrefslogtreecommitdiff
path: root/regress/usr.sbin
AgeCommit message (Collapse)Author
2019-11-06Cleanup code a bit, make sure the tal parse buffer is allways freed.Claudio Jeker
2019-11-04Reenable test-tal since problem is now fixed.Claudio Jeker
2019-11-04Adjust code to work with the new tal API functions (tal_read_file andClaudio Jeker
changed tal_parse).
2019-11-02Disable test-tal since it currently fails to build.Joel Sing
2019-11-02Use LibreSSL for rpki-client regress tests.Joel Sing
2019-10-03Now that we have snmp set, we can fully remove the dependency on netsnmp.Martijn van Duren
2019-09-20REGRESS_SKIP_TARGETS was spelled incorrectly.Alexander Bluhm
reported by Moritz Buhl
2019-09-18Let snmpd's regress test make use of our very own brand new SNMPv3/USMMartijn van Duren
support in snmp(1).
2019-09-17Rsyslogd TLS client side is unreliable. Disable compatiblity test.Alexander Bluhm
2019-09-10If rsyslogd is not installed, explicitly skip these syslogdAlexander Bluhm
compatibility tests. Fix some races and increase the timeout for rsyslog tests.
2019-09-10Fix more races in syslogd test.Alexander Bluhm
2019-09-10Fix race in syslogd test.Alexander Bluhm
2019-08-22Define the new verbose variable also in the regression sources soAlexander Bluhm
that the test programs link with the rpki-client object files again. from Moritz Buhl
2019-08-20shrink using new mechanismsMarc Espie
2019-08-20two simple scripts to shrink package generationMarc Espie
2019-08-20use wrapper module instead of rolling our ownMarc Espie
2019-08-20remove old cdrom=... propertyMarc Espie
2019-08-19silently add -DDONTLOGMarc Espie
2019-08-13Move comment about exabgp to the rule since this way it is shown to the user.Claudio Jeker
Reminded by bluhm@
2019-08-12Stop pulling libssl into rpki-client.Joel Sing
None of this code actually does TLS, hence libssl is not needed. Instead, pull in the correct headers and call the appropriate libcrypto initialisation functions (even this is only necessary to support OpenSSL prior to 1.1). While here also remove libssl/libcrypto initialisation/uninitialisation from main() - it should only be necessary in proc_parser(). ok deraadt@ job@
2019-08-09Move snmpd regress test from snmpctl(8) to snmp(1).Martijn van Duren
2019-08-07Actually make the hand coded ASPATH a AS_SEQUENCE and not an AS_SET.Claudio Jeker
2019-08-07The as0 regress can be enabled now. Needs exabgp to run else it is skipped.Claudio Jeker
2019-08-06Implement RFC7607 testing using exabgp since we need to fiddle withClaudio Jeker
attributes. Currently not enabled since it will fail.
2019-08-06Cleanup the scripts a bit, error out early if not run as root, useClaudio Jeker
set -x later in the script instead of ksh -x to shorten the output of the tests.
2019-08-06Update output file for mrt table-mp dumps since those can now dumpClaudio Jeker
the MPLS VPN as well.
2019-08-02pass -Dchecksum around so that checksum is exercised.Marc Espie
Also, inject fault and verify that package does indeed not get fully deleted
2019-07-17The Adj-RIB-Out is gone (or actually implemented differently). AdjustClaudio Jeker
the expected output files.
2019-07-16New realpath(3) requires existing files. Create fake disk imagesAlexander Bluhm
that are used in vmd(8) config parser.
2019-07-16so the output changed slightly, againMarc Espie
2019-07-12Test should also run if there is no obj directory. Name regressAlexander Bluhm
targets consistently with a run- prefix.
2019-07-11tweak tests to match simpler behaviorMarc Espie
also actually check the return code now that it makes sense
2019-07-10yet another creation instanceMarc Espie
2019-07-10actually we only care about the first error line, and the final statusMarc Espie
so just tell sed to do that
2019-07-10tweak the pipe to backsubstitute USER for `whoami` so that we can storeMarc Espie
the ref error messages
2019-07-10another error injectionMarc Espie
2019-07-09There was a race what is read first: the EOF from stdin to terminateAlexander Bluhm
the tty log collector, or log messages from the ptty. So some log messages could get lost at the end of testing. Now all available messages are read from the ptty before the test program terminates.
2019-07-09add a test that explicitly requires test to run.Marc Espie
thanks to tedu's tweaks to doas, there's no simpler way than to recurse thru make...
2019-07-09add a test using fault injection to trigger OpenBSD::Temp errorsMarc Espie
2019-07-09On slow armv7 machine reconnect timeout is faster than test, soAlexander Bluhm
error message appears twice.
2019-07-09don't look into installurlMarc Espie
2019-07-09a bit too enthusiastic in removing codeMarc Espie
2019-07-09byebye, superseded by genericMarc Espie
2019-07-09tweak the command runs to be able to inject code systematically, in orderMarc Espie
to trigger some bugs more easily
2019-07-06oops. added but not committed. Reminded by PatrickMarc Espie
2019-07-05Put some padding in Perl pack of struct timeval. On armv7 structAlexander Bluhm
timeval is 16 bytes, on i386 only 12 bytes were needed. The additional space makes setsockopt(2) succeed in armv7 regress. from Moritz Buhl
2019-07-04Add a dummy attr_writebuf() so that this code links again.Claudio Jeker
2019-07-04add a beautify test... which obviously found bugs, ahahMarc Espie
2019-06-28bgpct can now dump the BGP MPLS VPN entries as well. Adjust output file.Claudio Jeker
2019-06-28Remove some debug echosClaudio Jeker