Skip to content

chore: sync main with upstream splice 0.7.5 - #18

Merged
sadiq1971 merged 331 commits into
mainfrom
main
Aug 28, 2026
Merged

chore: sync main with upstream splice 0.7.5#18
sadiq1971 merged 331 commits into
mainfrom
main

Conversation

@sadiq1971

@sadiq1971 sadiq1971 commented Aug 27, 2026

Copy link
Copy Markdown
Collaborator

Catches this fork's main up with canton-network/splice:main. No code of ours in here.

  • Range: baf2552ef (2026-07-09) to b80adc142 (2026-08-27), 327 commits, 300 files
  • Version: 0.6.12 to 0.7.5
  • Our main had no fork-specific commits, so this is a straight catch-up rather than a merge of divergent work

OriolMunoz-da and others added 30 commits July 21, 2026 10:10
[static]

Signed-off-by: Oriol Muñoz <oriol.munoz@digitalasset.com>
… noise on CILR (#6482)

[static]

Signed-off-by: Mateusz Błażejewski <mateusz.blazejewski@digitalasset.com>
[ci]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
[ci]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
…on bumpting Canton (#6488)

[static]

Signed-off-by: Mateusz Błażejewski <mateusz.blazejewski@digitalasset.com>
[ci]

Signed-off-by: Mateusz Błażejewski <mateusz.blazejewski@digitalasset.com>
* [ci] bump nix alone

Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>

* [ci] fix all issues

Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>

* [ci] add package-lock

Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>

* [ci] chart timeout expanded

Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>

* [ci] update expected files

Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>

---------

Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>
On CILR we now need more than 5 minutes for `com.digitalasset.canton.sequencer.admin.v30.SequencerAdministrationService/OnboardingStateV2`.

The timing matches the switch to CantonBFT sequencers. Not fully clear how this can matter (CantonBFT only adds a ~constant amount of stuff to these, see https://github.com/DACH-NY/canton/blob/main/community/synchronizer/src/main/protobuf/com/digitalasset/canton/sequencer/admin/v30/sequencer_bft_additional_snapshot_info.proto).

I'm choosing to believe that we'll revisit the custom timeouts in DACH-NY/canton-network-internal#2125 some day so not bothering with deeper analysis.
Follow-up to #6453: the ACS growth alert now fires on splice_history_acs_snapshots_snapshot_size, but no dashboard showed it.

Also fixes the dashboard's template variables, which still queried the removed splice_store_acs_size metric.

[static]

Signed-off-by: Martin Florian <martin.florian@digitalasset.com>
[static]

Signed-off-by: Oriol Muñoz <oriol.munoz@digitalasset.com>
Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>
[ci]

Contains some cantonbft fixes (not yet enough for the next lsu).

Signed-off-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
Co-authored-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
To make GCP happy [static]

Signed-off-by: Oriol Muñoz <oriol.munoz@digitalasset.com>
Should avoid people forgetting an https:// in front.

fixes #6492

[ci]

Signed-off-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
Co-authored-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
Signed-off-by: Julien Tinguely <julien.tinguely@digitalasset.com>
Signed-off-by: Julien Tinguely <julien.tinguely@digitalasset.com>
And along the way fix some buggy scalatest clue strings.

fixes DACH-NY/cn-test-failures#9354

[ci]

Signed-off-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
Co-authored-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
* cantonbft - Adapt fork changes to include sequencer id in configured peers

[ci]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
Confirmed fix by clicking in the UI. Apparently this was broken ever since we hit serial / migration ID 10...

[static]

Signed-off-by: Martin Florian <martin.florian@digitalasset.com>
* feat: ensure pure gha can build images

Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>

* [ci] bump

Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>

---------

Signed-off-by: krzysztofczyz-da <krzysztof.czyz@digitalasset.com>
…ation cost (#6520)

[ci]

Signed-off-by: pasindutennage-da <pasindu.tennage@digitalasset.com>

Signed-off-by: Pasindu Tennage <pasindu.tennage@digitalasset.com>
[static]

Signed-off-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
Co-authored-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
Signed-off-by: Julien Tinguely <julien.tinguely@digitalasset.com>
This should avoid most of timeouts and starting new rounds which is quite a heavy operation with the downside of possible larger latency when some nodes are slow.

It will have a bigger impact if a node is down but we can handle that at an operational level.

[ci]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
Signed-off-by: Julien Tinguely <julien.tinguely@digitalasset.com>
[ci]

Signed-off-by: Oriol Muñoz <oriol.munoz@digitalasset.com>
* Extend LSU cancellation test to mimic actual cancellations

[ci]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
[ci]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
It doesn't make sense to add something to ablock after 30s actually

[ci]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
We already retry everywhere where we call that

This was interefering with the check for active synchronzier which is used in a lot of places and making things timeout because we had those implicit retries

In most places we had nested retries in this case

[ci]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
OriolMunoz-da and others added 16 commits August 26, 2026 09:49
…utting-down canton (#6958)

---------

Signed-off-by: Oriol Muñoz <oriol.munoz@digitalasset.com>
See DACH-NY/canton-network-internal#6371

Regular alerts for now, plan is to promote at least some of them to sev-2 and/or sev-1 after some testing.

- Discarded requests / replays <- also adds this as panels on the Global Synchronizer Utilization dashboard
- Failed confirmation requests
- TPS drop
* Expose sv app endpoints to the minimum users

Also extend open api so the other SVs can easily apply the same restriction

[static]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
[ci]

fixes #6970

Signed-off-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
Co-authored-by: Moritz Kiefer <moritz.kiefer@purelyfunctional.org>
Signed-off-by: Robert Autenrieth <robert.autenrieth@digitalasset.com>
Signed-off-by: Puneet Bharti <puneet.bharti@finoa.io>
Fixes #2336

Since #2336 was filed, #4113 deprecated `scan_txlog.py` and removed the code that ran it, but #5922 added a max tap amount of 100,000,000, so the existing test code failed with a "requirement ... was not met" error. This updates the test code to instead tap the max amount.

Signed-off-by: Matt Dziuban <mrdziuban@gmail.com>
Signed-off-by: Kajal <kajal.shah@digitalasset.com>
…6978)

---------

Signed-off-by: Oriol Muñoz <oriol.munoz@digitalasset.com>
* add the configs

Signed-off-by: Itai Segall <itai.segall@digitalasset.com>

* [ci] done

Signed-off-by: Itai Segall <itai.segall@digitalasset.com>

* [ci] fmt

Signed-off-by: Itai Segall <itai.segall@digitalasset.com>

* [ci] cleanup

Signed-off-by: Itai Segall <itai.segall@digitalasset.com>

* [ci] fix the test

Signed-off-by: Itai Segall <itai.segall@digitalasset.com>

* [ci]

Signed-off-by: Itai Segall <itai.segall@digitalasset.com>

---------

Signed-off-by: Itai Segall <itai.segall@digitalasset.com>
That warning doesn't exist in Canton anymore, see https://github.com/DACH-NY/canton/pull/30310/changes#diff-81c61bdb4b62b057717883fd710a5f662ee23d6769f03115d2db110dfda5829aL127

[ci]

Fixes #2706

Signed-off-by: Martin Florian <martin.florian@digitalasset.com>
---------

Signed-off-by: Oriol Muñoz <oriol.munoz@digitalasset.com>
)

And also the dashboard panel, and add a new panel that shows duplicate confirmation requests.

Follow-up to #6969

[static]

Signed-off-by: Martin Florian <martin.florian@digitalasset.com>
[ci]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
[ci]

Signed-off-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
Co-authored-by: moritz.kiefer@digitalasset.com <moritz.kiefer@purelyfunctional.org>
---------

Signed-off-by: Oriol Muñoz <oriol.munoz@digitalasset.com>
@sadiq1971 sadiq1971 changed the title feat: update main chore: sync main with upstream splice 0.7.5 Aug 27, 2026
[static]

Signed-off-by: Nicu Reut <nicu.reut@digitalasset.com>
Signed-off-by: Oriol Muñoz <oriol.munoz@digitalasset.com>
* add port 80 to helm VirtualServices

* direct GKE gateway to port 80 on istio

* also port 80 for cn-release-bundles and docs routes

* Cloud Armor got more restrictive about condition regexes

* fix sequencer gRPC ingress by forcing HTTP/2 in istio gateway chart

- found with docker-compose validator, doing this

    SEQ_HOST=sequencer-0.sv-2.$GCP_CLUSTER_HOSTNAME
    curl -sv --http2 \
        -H 'content-type: application/grpc' \
        -H 'te: trailers' \
        --data-binary $'\x00\x00\x00\x00\x00' \
        "https://${SEQ_HOST}/grpc.health.v1.Health/Check" \
        -o /dev/null

yields

    < HTTP/2 200
    < content-type: application/grpc
    < grpc-status: 14
    < grpc-message: upstream connect error or disconnect/reset before headers. reset reason: protocol error
    < server: istio-envoy
    < x-envoy-upstream-service-time: 3
    < via: 1.1 google

whereas with GKE L7 off we get

    < HTTP/2 200
    < content-type: application/grpc
    < grpc-status: 13
    < grpc-message: Half-closed without a request
    < x-envoy-upstream-service-time: 4
    < date: Mon, 24 Aug 2026 22:17:07 GMT
    < server: istio-envoy

---------

Assisted-by: Copilot:gpt-5.3-codex
Signed-off-by: Stephen Compall <stephen.compall@digitalasset.com>
Signed-off-by: Itai Segall <itai.segall@digitalasset.com>
@sadiq1971
sadiq1971 merged commit c8f3279 into canton-network:main Aug 28, 2026
4 of 7 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.