+ api/events resource
+ removing an autoattendant via /api/callforwards: events OK
+ removing an autoattendant via /api/cfdestiantions: OK
+ removing an autoattendant via /api/cfmappings: OK
Change-Id: I4c309753b9338582479dba9951f757bb2ecaad7e
+1. terminate if terminated contracts or subscribers exists
+2. if active contracts or subscribers -> error
+3. delete otherwise
+customer contact api DELETE
+customer contact testcases
+customer api POST
+customer api PUT/PATCH
+system contact api DELETE
+system contact testcases
+contract api POST
+contract api PUT/PATCH
+panel UI contact delete
+panel UI contract create
no change panel UI contract update
+panel UI customer create
no change panel UI customer update
note contact links like customers.omm_contact or customers.tech_contact
are not considered.
Change-Id: I19ea6e2245282485cc30d3e92d17a2eb603e0c2d
* I had to fix api-bannedusers.t otherwise it failed:
> ok 19 - bannedusers: check if all created test items have been foundin the list
> is_empty:1;ok 20 - fetch uri: https://10.15.20.74:1443/api/bannedusers/?page=1&rows=1 (OK: )
> $VAR1 = undef;
> ok 21 - Check that all added banned users were deleted 1..21 Failed 1/21 subtests
Change-Id: Ic9578c9f1ab20b2abeaa51656ed5556804228981
-undo unique lnp_numbers: undo misleading unique keys
-retain lnp_provider delete cascade restriction
-api: wire lnp_numbers_actual
-lnpnumbers collection pagination request takes several
seconds (up to tens of for 200k lnp numbers), too slow
especially for GET /api/lnpnumbers/?number=.. lookups
-query param is properly injected into the virtual view
sql query, so the lookup is fast now.
-panel UI custom datatable filter to switch between
complete history (all records) and actual portings
-"terminate" lnp number: DELETE /api/lnpnumbers/?number=x
+special use-case for melita's crm
+all empty or future "end" timestamps of records
are set to now(), so a subsequent listing shows no
records any more.
-number parameter is forced, other collection query params
are currently not supported for DELETE
Change-Id: I7b69d16a2d4f38ea84fc1cc0b407fa67e0f5ef8b
If set and a call is done within a pbx customer, the call is blocked
in blacklist, and is allowed in whitelist.
Change-Id: Ib43e5a45525a69e2b9dd9537dabddf10dfc006d1
File /.dockerinit no longer exists. With Docker versions 1.11 and
1.12 there's /.dockerenv available, though it's not yet sure
whether this will be present in the future too, so check for either
/.dockerenv or a proper cgroup setup inside the docker container.
File /proc/1/cgroup with Docker version 1.12.1 on Debian/jessie
looks like:
| root@4c3a1c6d773c:/# cat /proc/1/cgroup
| 8:freezer:/docker/4c3a1c6d773c1a5675f24ff58540ff1a290489e7325e1e66ec2c3392aa3a1145
| 7:perf_event:/docker/4c3a1c6d773c1a5675f24ff58540ff1a290489e7325e1e66ec2c3392aa3a1145
| 6:net_cls,net_prio:/docker/4c3a1c6d773c1a5675f24ff58540ff1a290489e7325e1e66ec2c3392aa3a1145
| 5:cpuset:/docker/4c3a1c6d773c1a5675f24ff58540ff1a290489e7325e1e66ec2c3392aa3a1145
| 4:blkio:/docker/4c3a1c6d773c1a5675f24ff58540ff1a290489e7325e1e66ec2c3392aa3a1145
| 3:cpu,cpuacct:/docker/4c3a1c6d773c1a5675f24ff58540ff1a290489e7325e1e66ec2c3392aa3a1145
| 2:devices:/docker/4c3a1c6d773c1a5675f24ff58540ff1a290489e7325e1e66ec2c3392aa3a1145
| 1:name=systemd:/docker/4c3a1c6d773c1a5675f24ff58540ff1a290489e7325e1e66ec2c3392aa3a1145
See https://github.com/docker/docker/issues/18355 and
https://github.com/docker/docker/pull/19490 for details.
Change-Id: I79fff0eead1a11f80b461791b2285a65d706262a
Also call agent request in the update_item method, to make it more safe.
And added lost transaction commit
Fix 500 response for the incorret x3_port input
Change-Id: Ic538ffdd4d7fff833ba5395934a1b2bc6cf71664
The codebase does not use autodie directly, it does indirectly via
Sipwise::Base, and the autodie module is part of the standard perl
distribution anyway. TryCatch is imported via Sipwise::Base, and
that depends on it. The rest are not used anymore in the codebase.
Change-Id: Ic08e246cc60d7d151381dff39e0eba517a492e06
+prohibit deletion of lnp providers if numbers are
still linked to it
+number+provider must be unique
+fixed other panel lnp page validation checks
+db constraints to stand concurrent inserts/updates
+fixup all txn_do die() calls in panel to properly
display the error
+bulk upload checks by db constraints
+lnp api testcase
Change-Id: I872b19f50d96e6c39c7fd5728b2a16555a397236
- the connector name is properly set when creating a reseller
- the network tag defaults to just (sip,xmpp,webrtc,...)
- rtcsessions doesn't accept network tag anymore
- an rtc-account is created for every network of the rtc-user
- the identifier in the rtc-account is set according to the specific connector
- api-rtc-full.t has been simplified for the user (no more need for domain_id)
- update available networks (in reseller api)
Change-Id: I2700c646726a06569cc5a70a5a30d4cb67eb1c80
+creation of profile packages with hour and minute
intervals was disabled in api-balanceintervals.t
Change-Id: Idd18ced9704522bf5d919155d866c91b81526b92
If no subscriber existed, first is a pilot, and doesn't consider groupmembers.
Also fix getting created hal, which was unconditionally overwritten by loaded
Change-Id: Ie2cf76624f125dbc9aaffc1ffab19319028c2c6f
by choosing "stable" (about 1m30s) or "fast" (about 30s) as second
argument of testrunner, only tests are run, which are currently
succeeding on a plain trunk vm using docker. (times are on my machine)
I hope we can in this way re-enable them on nightlies (or even in gerrit)
so that devs have some better feedback on the impact of our changes.
Current status of tests (in trunk):
| Problems:
| api-callforwards.t
| api-balanceintervals.t
| api-faxes.t
| api-journals.t
|
| Fixed:
| api-all-links.t
| api-contracts.t
| api-customercontacts.t
| api-customers.t
| api-rewriterulesets.t
| api-root.t
|
| Working:
| api-billingfees.t
| api-billingnetworks.t
| api-billingprofiles.t
| api-billingzones.t
| api-calllists.t
| api-calls.t
| api-cert-auth.t
| api-cfdestinationsets.t
| api-ncoslevels.t
| api-pbxdevicemodels.t
| api-pbxdevices.t
| api-peeringgroups.t
| api-peeringrules.t
| api-peeringservers.t
| api-preferences.t
| api-profilepackages.t
| api-resellers.t
| api-rewriterules.t
| api-soundsets.t
| api-subscriberregistrations.t
| api-subscribers.t
| api-systemcontacts.t
| api-threads.t
| api-topuplogs.t
| api-trustedsources.t
| api-valid-patch.t
| api-vouchers.t
Change-Id: I74e672d9dc6bcccfd3ace2fe233dbcf325c6fc4d
+input validation for verlapping time ranges
for both weekdays and special. can be disabled.
+root entiy locking
+peektime special is growing and not paginated,
so PUT/PATCH will get slower if special peektimes
are added on a regular basis. they can be removed
using the UI however
+testcase to check overlap detection
Change-Id: I935d943078ab5c81263da88ecd04e004deb26c8b
Otherwise we are failing with error:
> t/api-rest/api-faxes.t ..
> ok 1 - check faxserver feature state: disabled
> ok 2 - check error message in body: Faxserver feature is not active
> # Tests were run but no plan was declared and done_testing() was not seen.
> Dubious, test returned 254 (wstat 65024, 0xfe00)
> All 2 subtests passed
>
> Test Summary Report
> -------------------
> t/api-rest/api-faxes.t (Wstat: 65024 Tests: 2 Failed: 0)
> Non-zero exit status: 254
> Parse errors: No plan found in TAP output
Change-Id: Iff32b25ced931931af8cbaedc7fb054e48773667
Otherwise we are getting error:
> not ok 10 - Check existence of rating_status field
>
> # Failed test 'Check existence of rating_status field'
> # at t/api-rest/api-calllists.t line 50.
> # Total number of calls: 0
Change-Id: I257e198b740363d5d25d6d35a29a482f966377fb
+subscriber instead of source_subscriber typo fixed
+rudimentary test case to see if the filter works
+fixed in/out direction filter testcase: sum of in and
out calls should be >= total count of calls
Change-Id: Ibe4b8153f023c53bb463626ad9e6613992fc101d
+basically, the fix means to set
billing.contract_balances.start
billing.contract_balances.end
billing.contracts.create_timestamp
billing.contracts.modify_timestamp
to the 'local' timezone.
+test cases added
Change-Id: I203f8fc93df2b53c1cfe89bb1a90cdf6388e20b3
Fix faxserversettings PUT/PATCH processing found after faxes tests run
MT#15441 Tests for subscriber number changes
MT#12347 Test framework small fixes and additions
Some format changes and unused library removals
Add put_and_get method as common
Replace request_get & check 200 with check_item_get
Changes for API that doesn't return created object json after POST
Complete base_uri using replacement with normalize_uri
Change-Id: Ia31a97eb9511f1ca1de63c72bcbe380b89dfed34
any *attempt* of a top-up request should be
logged.
+create top-up log records for api
in order to record failed top-up requests as well,
the somewhat tricky thing is to have two
separate transactions here
+propery casting of topup request json field
values to numbers etc., so the transaction for
creating the log entry cannot fail and e.g.
requests with subscriber_id='blah' are recorded
correctly
+new "request_token" parameter for /api/topupvoucher
and /api/topupcash, to identify and filter for
particular requests.
+topup log api resource
+topup log api tests
+fix to correctly reject used vouchers
+topup log panel UI
+fix for balanceintervals.t threaded tests
Change-Id: I86eb845f6173803705b12cc7e5cdbac9a3153a0a
+refactoring recent failing api-balanceintervals.t test to
need only two concurrent transactions (instead of three,
see https://bugtracker.sipwise.com/view.php?id=15277
Change-Id: I4902e616c3adfae81676194fa5e2b99e4b516ed1
Irina Peshinskaya
Rene Krenn
Alexander Lutay
Andreas Granig
Gerhard Jungwirth
Guillem Jover
Kirill Solomko
Victor Seva
Michael Prokop