Repository navigation
nomad job dispatch gives "failed to create result paginator" with 1.11.3 client against 1.11.2 server #27680
Description
Nomad version
1.11.2 (server) and 1.11.3 (client)
Operating system and Environment details
Linux redacted 6.1.0-43-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.162-1 (2026-02-08) x86_64 GNU/Linux
Issue
When dispatching a job from the CLI, using a 1.11.3 client towards a 1.11.2 server, we get the following error:
$ docker run --rm hashicorp/nomad:1.11.3 job dispatch -address=http://example.com:4646 dsconfig
Error querying job prefix "dsconfig": Unexpected response code: 400 (failed to create result paginator: failed to read filter expression: 1:44 (43): no match found, expected: "empty" or [ \t\r\n])
Expected Result
With a 1.11.2 client we get (without auth)
$ docker run --rm hashicorp/nomad:1.11.2 job dispatch -address=http://example.com:4646 dsconfig
Failed to dispatch job: Unexpected response code: 403 (Permission denied)
or with auth
$ docker run --rm hashicorp/nomad:1.11.2 job dispatch -token=redacted -address=http://example.com:4646 dsconfig
Dispatched Job ID = dsconfig/dispatch-1773660299-50ff1563
Evaluation ID = a5d52258
...
Actual Result
$ docker run --rm hashicorp/nomad:1.11.3 job dispatch -address=http://example.com:4646 dsconfig
Error querying job prefix "dsconfig": Unexpected response code: 400 (failed to create result paginator: failed to read filter expression: 1:44 (43): no match found, expected: "empty" or [ \t\r\n])
Activity
tgross commentedon Mar 16, 2026
Oops, this one is on me, originating from: #27631. This will also impact the
nomad job periodic forcecommand. The server-side is getting ago-bexprexpression it doesn't understand, so it can't construct the paginator.Sorry about that, we'll get it on the roadmap to fix up. The workaround for the meantime would be to use a CLI client that's no newer than the server. (Internal ref https://hashicorp.atlassian.net/browse/NMD-1296)
eyupcanakman commentedon May 16, 2026
I'd like to take this one. I reproduced it locally: a 1.11.3 CLI against a 1.11.2 server fails on
job dispatchandjob periodic force. #27631 switched the prefix lookup to a server-side go-bexpr filter usingis not nil, and older servers run a go-bexpr version without that operator, so they reject the filter with a 400.I'm thinking: keep the server-side filter, and when an old server rejects it, fall back to the old prefix list call and filter client-side so newer servers still get the #27631 load reduction. Does that direction look OK, and is NMD-1296 already being worked on so I don't duplicate it?
tgross commentedon May 18, 2026
That seems like a reasonable approach @eyupcanakman. For permissions-related reasons we can't assign an issue to you directly. But it's not being worked on currently, so feel free to take a crack at it.
cli: fix job dispatch and periodic force against older servers
cli: fix job dispatch and periodic force against older servers (#28014)
cli: fix job dispatch and periodic force against older servers (#28014)
Backport of cli: fix job dispatch and periodic force against older se…
cli: fix job dispatch and periodic force against older servers (#28014)
cli: fix job dispatch and periodic force against older servers (hashi…
github-actions commentedon Sep 30, 2026
I'm going to lock this issue because it has been closed for 120 days ⏳. This helps our maintainers find and focus on the active issues.
If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.