This Patroni log means its etcd member-discovery lookup raised an exception; the message alone does not identify why. An Ubuntu 24.04 report points to an etcd API response problem, but that account could not be independently verified, and a separate Patroni issue reports the same exception with etcd 3.4.36. Check the versions and the response from the exact endpoint Patroni contacts before changing packages or DCS settings.
What the error tells you—and what it does not
A Patroni issue opened June 5, 2026 reports this log while using a configuration with etcd3: and no etcd: section:
As an Amazon Associate I earn from qualifying purchases.
Failed to get list of machines from http://10.245.7.12:2379/v3: AttributeError("'int' object has no attribute 'get'")
The reporter listed Patroni 4.1.3, Python 3.12, etcd 3.4.36, and PostgreSQL 18. That is one reported environment, not proof that every installation with those versions—or Ubuntu 24.04 generally—has the same fault. See the Patroni issue report.
#1 Best Overall
Patroni’s source defines _get_machines_list, calls a member lookup for candidate base URIs, and logs this message if an exception occurs. That ties the log to member discovery, but does not establish the malformed response, API incompatibility, or configuration condition behind a particular occurrence. The code path can be inspected in Patroni’s etcd DCS source.
Why Ubuntu 24.04 is mentioned
An October 1, 2026 DEV Community article attributes a similar failure in its test setup to Ubuntu 24.04’s etcd 3.4.30 package. Its search-result text says a POST /v3/cluster/member/list request returned 404 on etcd 3.4.30 and a member list on etcd 3.6.15, and reports that upgrading worked in that test. The page itself could not be retrieved for independent verification, so these are the author’s claims, not confirmed general compatibility guidance.
The separate issue report involves etcd 3.4.36, not 3.4.30. That difference is another reason not to diagnose a machine from its Ubuntu release alone. Determine which etcd binary and package are actually running, then inspect what the configured endpoint returns.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →Diagnose the failure from the Patroni host
- Record the running versions and configuration. Check the Patroni, Python, and etcd versions, including etcd on every member. Record the active DCS block and the client URLs Patroni uses. Confirm the installed package and running binary rather than assuming Ubuntu’s bundled version is in use.
- Query the endpoint Patroni actually contacts. From the Patroni host, capture the HTTP status and response body for the relevant member-discovery request. The Ubuntu article proposes
POST /v3/cluster/member/listwith an empty JSON body, but verify that endpoint and request format for the versions you run before treating it as a universal test. Account for any proxy, TLS, or address-routing behavior between Patroni and etcd. - Compare the response with the configured DCS backend. Check that Patroni is pointed at the intended etcd client endpoint and that the response is consistent with its configured backend. The reported error occurred with
etcd3:alone; do not add anetcd:block or switch backends just because the exception text mentions a machine list. - Use the observed failure to choose the next step. If the endpoint is missing or returning an incompatible response, check the official release and compatibility documentation for the exact Patroni and etcd versions, then reproduce the issue in a controlled environment. A successful health check is not enough to establish that Patroni’s member-discovery request succeeds: the checks exercise different behavior.
When an etcd upgrade may be appropriate
Consider an upgrade only after the endpoint response and installed versions support that diagnosis. The Ubuntu-specific report recommends moving from etcd 3.4 to a newer upstream release and says 3.6.15 worked in its setup, but neither claim establishes a universal fix or a supported path for every cluster.
For an existing cluster, consult etcd’s official upgrade and backup guidance for the exact starting and target releases before changing binaries. Do not wipe data directories or follow a generic sequence of upgrades based on the unverified article; the correct procedure depends on the cluster’s current state and supported upgrade path. Test the change in a controlled environment and confirm Patroni discovery recovers before applying it to production.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Should you switch to the legacy etcd backend?
Not on the basis of this error alone. The reported reproduction used etcd3: without an etcd: section, and the exception does not demonstrate that switching backends would fix the underlying response. Before changing DCS configuration, verify official guidance for the exact Patroni and etcd releases and assess the effect on the deployment.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
