Bug #5967
closedBTS_Tests_OML.TC_wrong_trx_nr fails
100%
Description
BTS_Tests_OML.TC_wrong_trx_nr fails.
Description: test behavior for wrong TRX number in object instance
error: Timeout waiting for NACK
Looks like osmo-bts is not sending an error message (NACK) in situations where it should?
Files
Related issues
Updated by arehbein 2 months ago
- Status changed from New to In Progress
- % Done changed from 0 to 40
I uploaded some patches (topic osmo_bts_nack).
They all lead up to the last one https://gerrit.osmocom.org/c/osmo-bts/+/32076 which I have marked as WIP, because I'm not certain all of the causes set are correct.
Any corrections on what causes would be correct in case any are wrong would be appreciated. Plus maybe some tests will break (for good or bad), because we transition from the generic NM_NACK_OBJINST_UNKN (not sure when that is supposed to be used - when is the object unknown.. or is this just a very generic cause? - similar for the other enum constants...)
Updated by arehbein 2 months ago
- File configure.txt configure.txt added
I tried enabling all the subprojects, but configure fails, I've attached the ./configure message (the last line is not cut off, that's how it appeared).
The command was (not sure if that's allowed to work even)osmo-bts$ ./configure --enable-trx --enable-oc2g --enable-octphy --enable-sysmocom-bts --enable-litecell1
Updated by fixeria 2 months ago
arehbein wrote in #note-2:
I tried enabling all the subprojects, but configure fails, I've attached the
./configuremessage (the last line is not cut off, that's how it appeared).The command was (not sure if that's allowed to work even)
osmo-bts$ ./configure --enable-trx --enable-oc2g --enable-octphy --enable-sysmocom-bts --enable-litecell1
You don't really need all these BTS backends to test the common OML part, which they all share. In ttcn3-bts-test we're executing the testcases against osmo-bts-trx, so --enable-trx should be enough for testing your patches. It does not require any vendor specific headers.
Updated by pespin about 2 months ago
- Related to Bug #5966: BTS_Tests_OML.TC_wrong_obj_class fails added
Updated by arehbein about 2 months ago
fixeria wrote in #note-3:
arehbein wrote in #note-2:
I tried enabling all the subprojects, but configure fails, I've attached the
./configuremessage (the last line is not cut off, that's how it appeared).The command was (not sure if that's allowed to work even)
osmo-bts$ ./configure --enable-trx --enable-oc2g --enable-octphy --enable-sysmocom-bts --enable-litecell1You don't really need all these BTS backends to test the common OML part, which they all share. In ttcn3-bts-test we're executing the testcases against osmo-bts-trx, so
--enable-trxshould be enough for testing your patches. It does not require any vendor specific headers.
In this case, I just wanted to make sure compilation is done on all components in order to check if there are compiler error. Maybe it's better to have Jenkins try out 'fringe' compilation cases
Updated by arehbein about 2 months ago
- % Done changed from 40 to 80
These should constitute a fix (if not, then I'll probably have to check if something is wrong with oml_fom_ack_nack()):
https://gerrit.osmocom.org/c/osmo-bts/+/32354/2
https://gerrit.osmocom.org/c/osmo-bts/+/32075/6
Can't run it through Jenkins yet myself; there was a change to docker-playground a month or so ago that broke my customization scripts with which I would usually try out integration tests locally.
Updated by arehbein about 1 month ago
- Status changed from In Progress to Resolved
- % Done changed from 80 to 100
Applied in changeset osmo-bts|cf8736e9160315c3d85b97f4b498be1605f1f129.
