resolver-benchmarking merge requestshttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests2023-11-24T14:47:47+01:00https://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/82Update resource-monitor version2023-11-24T14:47:47+01:00Tom KrizekUpdate resource-monitor version... to allow !81 monitoring the `http_urls` endpoints... to allow !81 monitoring the `http_urls` endpointshttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/81Allow customizing resource-monitor config2023-11-24T14:47:32+01:00Tom KrizekAllow customizing resource-monitor configUse a default resmon.yaml config from resource-monitor repo and
optionally turn on BIND statistics channel monitoring.Use a default resmon.yaml config from resource-monitor repo and
optionally turn on BIND statistics channel monitoring.https://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/79host_vars: remove non-existent machines2023-09-27T09:07:33+02:00Oto Šťávahost_vars: remove non-existent machineshttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/76Add ansible-lint to CI2023-09-26T13:31:13+02:00Oto ŠťávaAdd ansible-lint to CIhttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/78Namespace dnsperf input variables2023-09-26T13:29:19+02:00Petr Špačekpspacek@isc.orgNamespace dnsperf input variablesFixes last ansible-lint var-naming[no-role-prefix] complaints.Fixes last ansible-lint var-naming[no-role-prefix] complaints.https://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/77Fix perf role2023-09-26T13:25:50+02:00Petr Špačekpspacek@isc.orgFix perf rolehttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/75Ansible cleanup round 22023-09-26T10:45:09+02:00Petr Špačekpspacek@isc.orgAnsible cleanup round 2https://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/73Ansible Cleanup2023-09-20T16:11:45+02:00Petr Špačekpspacek@isc.orgAnsible CleanupPreparatory cleanup before commencing #14. For now I'm trying to make ansible-lint happy.
Except for Condor removal this MR does not change semantics and should not have material
effect on execution.
References: #14Preparatory cleanup before commencing #14. For now I'm trying to make ansible-lint happy.
Except for Condor removal this MR does not change semantics and should not have material
effect on execution.
References: #14https://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/74Add new bind dependencies and customization options2023-09-19T16:46:30+02:00Tom KrizekAdd new bind dependencies and customization optionsI've been juggling these changes in `isc-shotgun` branch for a while now since some of these changes have been in development. Now these have been stabilized, there's no need to carry them in a separate branch.I've been juggling these changes in `isc-shotgun` branch for a while now since some of these changes have been in development. Now these have been stabilized, there's no need to carry them in a separate branch.https://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/72Filter-out main and child thread machinery from perf output2023-08-01T17:27:01+02:00Ondřej SurýFilter-out main and child thread machinery from perf outputThe main and child threads have a slightly different function trace in
the perf output that prevent combining them together.
Filter-out the differences between the two, so there's one unified
combined output from the flamegraph.The main and child threads have a slightly different function trace in
the perf output that prevent combining them together.
Filter-out the differences between the two, so there's one unified
combined output from the flamegraph.Ondřej SurýOndřej Surýhttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/71BIND: libcap is now mandatory2022-12-08T16:26:55+01:00Petr Špačekpspacek@isc.orgBIND: libcap is now mandatoryhttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/70shotgun: add extra scenarios2022-08-29T10:44:34+02:00Tom Krizekshotgun: add extra scenariosA few additional shotgun configs (scenarios) that can be used for testing.A few additional shotgun configs (scenarios) that can be used for testing.Oto ŠťávaOto Šťávahttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/69bind: parametrize additional named.conf options2022-08-22T07:57:36+02:00Tom Krizekbind: parametrize additional named.conf optionsThe only change here is the increase to 10k recursive clients from the default 1k (https://bind9.readthedocs.io/en/v9_18_5/reference.html#namedconf-statement-recursive-clients)The only change here is the increase to 10k recursive clients from the default 1k (https://bind9.readthedocs.io/en/v9_18_5/reference.html#namedconf-statement-recursive-clients)Oto ŠťávaOto Šťávahttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/68Support pulling shotgun/flamegraph images from registries2022-08-05T07:41:43+02:00Tom KrizekSupport pulling shotgun/flamegraph images from registriesThis enables pulling the flamegraph and shotgun images from registries.
---
Most of this MR changes are just indentation -- use `Hide whitespace changes` for the actual changes, which are quite brief.
@ostava This should be a no-op fo...This enables pulling the flamegraph and shotgun images from registries.
---
Most of this MR changes are just indentation -- use `Hide whitespace changes` for the actual changes, which are quite brief.
@ostava This should be a no-op for your existing workflow, as all the existing fallbacks to build the image locally are in place.https://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/67resource monitor: collect resource usage data during test2022-07-27T11:27:28+02:00Tom Krizekresource monitor: collect resource usage data during test@ostava This role is completely optional and is turned ~~on~~ off by default to not mess with your setup.
In case you're interested, it can be utilized to generate [pretty neat charts](https://indico.dns-oarc.net/event/43/contributions/...@ostava This role is completely optional and is turned ~~on~~ off by default to not mess with your setup.
In case you're interested, it can be utilized to generate [pretty neat charts](https://indico.dns-oarc.net/event/43/contributions/913/attachments/888/1620/pspacek-performance-effects-of-dnssec-validation.pdf) with the help of some other [scripts](https://gitlab.isc.org/isc-projects/resource-monitor).https://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/66shotgun: ensure generator timestamps are integers2022-07-27T09:30:14+02:00Tom Krizekshotgun: ensure generator timestamps are integersThe conversion into integer should happen after the division, otherwise
we end up with a float.
Additionally, the condition was problematic in Ansible 6.0+ and since it
didn't make much sense anyway, it was removed. It doesn't seem to b...The conversion into integer should happen after the division, otherwise
we end up with a float.
Additionally, the condition was problematic in Ansible 6.0+ and since it
didn't make much sense anyway, it was removed. It doesn't seem to be
possible to set facts and check their value in the same task.https://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/65various updates2021-12-22T16:18:07+01:00Tomas Krizekvarious updateshttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/64ansible: support for collapsible job logs format for Gitlab2021-09-13T11:14:49+02:00Petr Špačekpspacek@isc.organsible: support for collapsible job logs format for Gitlabhttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/63bind: ability to cache non-tagged images2021-07-22T13:38:37+02:00Petr Špačekpspacek@isc.orgbind: ability to cache non-tagged imageshttps://gitlab.nic.cz/knot/resolver-benchmarking/-/merge_requests/62roles/shotgun: use v20210714 release2021-07-20T14:24:12+02:00Tomas Krizekroles/shotgun: use v20210714 release