nano v2.0.31
nano v2.0.31: a hotfix. Fresh role-split installs no longer return 503 when a slow ClickHouse health check fails, scripts/nano upgrade no longer fails its SIEM Health check on errors the rollout itself caused, and the installer's re-run upgrade prompt defaults to yes.
v2.0.31 includes everything in v2.0.30. It is a hotfix with no security fixes and no new migrations.
To upgrade, follow Upgrade a self-hosted deployment and the Upgrade notes below. v2.0.20 is a required stop: installs on a release older than v2.0.20 upgrade to v2.0.20 first.
Fixes
- Role-split installs no longer return 503 after one slow ClickHouse health check. On a fresh role split with one ClickHouse node, the app servers' ClickHouse proxy could mark that node down after a single slow check. With no server left, every app request to ClickHouse got
503, and enabling Agent Lane failed. The proxy now allows a 5 second check and needs three failures in a row to mark a server down. A backend with only one server also gets an unchecked backup entry for the same node, so a check flap never empties it. A node that is really down still returns its own connection error, and SIEM Health still shows it down. Backends with several servers still fail over, now within about 5 seconds of a node going down. - Agent Lane retries briefly. Creating Agent Lane users and minting tokens now retries a proxy
503or a network error up to 3 times over about 5 seconds. A ClickHouse error is never retried. scripts/nano upgradeno longer fails on errors the upgrade itself caused. The final SIEM Health check could fail with "ClickHouse is hitting its memory limit" because the rollout reloads dictionaries, and--resumecouldn't pass for a day after. Now a new ClickHouse memory-limit or dictionary-update error doesn't fail the upgrade if it wasn't seen in the 24 hours before the upgrade, happened only during the rollout, and has been quiet for 5 minutes (NANO_UPGRADE_TRANSIENT_ERROR_QUIET_SECONDS). It is listed in the upgrade summary instead. Errors that continue, or that were there before, still fail it.--resumegets a fresh SIEM Health report. After a failed SIEM Health check,scripts/nano upgrade --resumeused to reuse the same report, which is only made every 12 hours. It now restarts nano-jobs, which runs a new report, and waits up to 10 minutes (NANO_UPGRADE_SIEM_HEALTH_WAIT_SECONDS) for it.- The re-run upgrade prompt defaults to yes. A
./install.shre-run in a terminal that changes release now asks[Y/n]: Enter upgrades.nstops with exit code3, and end of input with exit code4; nothing is changed.NANO_CONFIRM_UPGRADE=1and runs with no terminal are unchanged. - Installer wording for Vector native senders. A role-split ingest server reports
1 sender anchoror0 sender anchors from the shared .env, and the port6000mutual TLS notice only shows on an upgrade from a plaintext listener, never on a fresh install.
Upgrade notes
- Required stop at v2.0.20. See above.
- From v2.0.30, there is nothing extra to do. On a role split,
scripts/nano upgraderenders the new ClickHouse proxy configuration on each app server. - From v2.0.29 or earlier, also read the v2.0.30 upgrade notes: port
6000requires mutual TLS from v2.0.30. - If you script
./install.shre-runs through a terminal, Enter now upgrades. SetNANO_CONFIRM_UPGRADE=1to make that explicit, or answernto stop.
nano v2.0.32
nano v2.0.32: Hunts for Self-run, pivt page awareness with actions you confirm, dataset=signals in Search, queries shown as nPL, no lost events across upgrades and backups, and an announced upgrade path for open 0.1.702 OCSF installs.
nano v2.0.30
nano v2.0.30: Vector native port 6000 always requires mutual TLS, with a certificate per sender. The installer checks the host before it asks anything, shows a review screen, and Self-run installs and upgrades log in to the registry themselves.