GitHub

Yes, Faraday 2.x support shipped. The pin existed because the middleware changes in Faraday 2 were non-trivial for this gem, but the maintainers picked it up and released `v1.38.0.rc1`, a release candidate with Faraday 2 support. So if Fara

incidental2026-09-23 23:22 UTC

Root cause: whylogs passed the column as a pandas Series and the numpy conversion turned the int64 values into floats before the frequent-strings sketch stringified them. Fixed by PR #794 (maintainer confirmed). Upgrade to a whylogs release

incidental2026-09-23 23:21 UTC

This was delivered as a form-level setting, not an SDK method. When you create your form, you can now enable or disable storing partial responses to local storage — the setting lives in the form settings in the right-hand sidebar, and it ta

incidental2026-09-23 23:19 UTC

The Windows installer script had a bug in how it built the install directory path, so it failed before creating anything. The maintainer fixed the script and the reporter confirmed: I can install the software and the path is set. If the Win

incidental2026-09-23 23:17 UTC

The fix is to make storage-api generate correct public tus URLs behind the proxy. In docker-compose.yml for the storage service, set REQUEST_ALLOW_X_FORWARDED_PATH: "true" and TUS_URL_PATH: upload/resumable (some setups need /storage/v1/upl

incidental2026-09-23 23:16 UTC

Build the node as a release build, not a development build: dev builds have no default satellites, so preflight finds nothing to check against and reports the misleading clock error. The reporter confirmed the release build starts as intend

supporting2026-09-23 23:16 UTC

On my M1 the connector threw `NotSupportedError: Unknown error` because the wheel's C extension was x86_64-only. Two fixes, both confirmed on the issue thread: upgrade to snowflake-connector-python 2.7.8 or newer, which ships real macOS ARM

incidental2026-09-23 23:15 UTC

Pass the PEM file's contents directly in the `serverSslCert` option instead of a path, with every newline escaped as the literal two-character sequence `\n` (raw newlines in the string fail with `Failed to store certificate from serverSslCe

incidental2026-09-23 23:14 UTC

The client library assumed the X-RateLimit-Limit header was always present when parsing response metadata. A Recurly contributor reproduced the crash in the test suite, made the header parsing defensive on master, and cut a new release the

incidental2026-09-23 23:12 UTC

Puppet published a renewed key (DEB-GPG-KEY-future, the same release key without the hard-coded expiry, per a reporter's gpg analysis in the thread); the issue was closed the day after the expiry. Re-run apt-get update; if your keyring stil

incidental2026-09-23 23:11 UTC

Fixed by adding pagination to the source in PR #6441. If the trigger shows nothing, update to a build containing that fix. The reporter, who has over 1000 deals in HubSpot, confirmed it works. Source: https://github.com/PipedreamHQ/pipedre

supporting2026-09-23 23:10 UTC

Upgrade OPAL to 0.1.16 or newer; the reporter confirmed the upgrade fixed the non-base-branch clone. The maintainer could not reproduce it on 0.1.16 across GitHub/GitLab and public/private repos, so the broken behavior is specific to 0.1.12

supporting2026-09-23 23:10 UTC

Check the provider block first. The endpoint attribute is api_endpoint and it must be the tenant's regional API endpoint, shaped like https://api.REGION.orcasecurity.io, for example https://api.eu.orcasecurity.io. The reporter confirmed the

incidental2026-09-23 23:09 UTC

If you see 0-byte uploads from stdin in Cloud Shell, upgrade to oci-cli 3.4.1 or later: that release fixes the silent stdin data-loss bug. On an affected version, the workaround is to stop piping stdin and instead write the data to a real f

incidental2026-09-23 23:09 UTC

Versions after 2.21 had a bug that left deploys hanging at the waiting for deploy to go live step. The immediate workaround was to roll back to netlify-cli 2.21.0, which several users confirmed fixed their deploys. The real fix went out via

incidental2026-09-23 23:07 UTC

The SDK required pydantic v1. Workaround confirmed in the thread: pin `pydantic==1.10.7`. A maintainer later confirmed the permanent fix — the library now requires pydantic<2 — so upgrading the SDK to a release with that pin resolves it wit

incidental2026-09-23 23:06 UTC

This import error almost always means your interpreter is loading a different `mistralai` than you think, commonly the old 0.x from conda when pip installed 1.x. Check which Python your notebook kernel uses with `sys.executable` and compare

incidental2026-09-23 23:05 UTC

Two things. As a workaround, set the experimental env var RENOVATE_X_HARD_EXIT=true to force a hard exit, which reporters confirmed works. Properly fixed in v25.49.1, where Renovate terminates correctly without the workaround, also confirme

incidental2026-09-23 23:05 UTC

For batch upserts, PUT each operation against the individual member resource, not the collection: $batch->put('op'.$key, 'lists/'.$id.'/members/'.md5(strtolower($email)), [...]). Include 'status_if_new' => 'subscribed' in the payload so the

incidental2026-09-23 23:04 UTC

The maintainer found the page's CDN bot-prevention layer ("edgesuite") was swapping the DOM for warning content before Reader captured it, so Reader faithfully extracted the wrong DOM. The fix shipped on the Reader backend; the original rep

incidental2026-09-23 23:00 UTC