inetshell

Grafana with SSO: logging in with the local user (?disableAutoLogin), roles overwritten on every login (skip_org_role_sync) and provisioned data sources with secrets

/login?disableAutoLogin=true to reach the local form; grafana cli admin reset-admin-password if you also forgot the password. skip_org_role_sync: true per provider so Grafana stops overwriting roles; or role_attribute_path with role_attribute_strict. Data sources in provisioning/datasources with editable: false, secrets in secureJsonData, removal with deleteDatasources or prune: true.

Grow a Linux VM disk online: rescan, growpart, pvresize, lvextend and xfs_growfs

echo 1 > /sys/class/block/sda/device/rescan; growpart /dev/sda 3; pvresize /dev/sda3; lvextend -r -l +100%FREE /dev/vg/lv. Without LVM: resize2fs or xfs_growfs. All online.

helmfile in production: three-way diff, 'already set in map', OCI 403, duplicate dependencies and adopting resources that were already deployed

Three variables (HELM_DIFF_THREE_WAY_MERGE, HELM_DIFF_NORMALIZE_MANIFESTS, HELMFILE_GOCCY_GOYAML), two flags (–skip-diff-on-install, -l name=) and one rule: pin helm, helmfile and helm-diff in CI. Adopting existing resources = the managed-by=Helm label plus the meta.helm.sh/release-name and release-namespace annotations.

Importing dumps into Cloud SQL without errors: DEFINER and SUPER (ERROR 1227) on MySQL, extensions and owners on PostgreSQL

MySQL: mysqldump –hex-blob –single-transaction –set-gtid-purged=OFF –default-character-set=utf8mb4; DEFINER clauses for users that do not exist on the instance (or root@localhost) break the import. PostgreSQL: pg_dump –format=plain –no-owner –no-acl plus sed to comment out DROP/CREATE/COMMENT ON EXTENSION. Instance service account with objectAdmin on the bucket; gcloud sql import sql with –database and –user.

Mount a qcow2 disk on the host with qemu-nbd (read or repair files of a powered-off VM)

modprobe nbd max_part=16; qemu-nbd -c /dev/nbd0 vm.qcow2; partprobe; mount /dev/nbd0p1. When done, umount and qemu-nbd -d, or the VM will not start because of the lock. VM off, always; -r to only read.

Nextcloud: 'will not be accessible due to incompatible encoding' (files invisible because of NFD or Latin-1 names)

convmv -f utf-8 -t utf-8 -r –notest –nfc and then occ files:scan –all. Without –notest it only simulates. For Latin-1 names, convmv -f iso-8859-1 -t utf-8. From macOS, rsync –iconv=utf-8-mac,utf-8.

Oracle CQN with Python: learn about table changes without polling (and why oracle_fdw fires no triggers)

GRANT CHANGE NOTIFICATION; oracledb.connect(events=True); conn.subscribe(callback, operations, qos); sub.registerquery(‘select id, status from orders’). No polling. The database must be able to connect to your app (or use client_initiated=True with Instant Client 19.4+).

pgbouncer: 'unsupported startup parameter: extra_float_digits' (and 'no such database')

ignore_startup_parameters = extra_float_digits, options in pgbouncer.ini and RELOAD. extra_float_digits can be ignored with no consequences; ‘options’ (search_path) cannot: set it on the role with ALTER ROLE SET. ’no such database’ means the entry is missing from [databases] or the * wildcard.

PostgreSQL: admin cheat sheet (rows per table, sizes, sessions, read-only user)

Exact per-table count with the xpath trick, estimate with pg_stat_user_tables, pg_size_pretty for sizes, pg_terminate_backend or DROP DATABASE WITH (FORCE) for sessions, and GRANT pg_read_all_data for the reporting user.

Removing Rancher from a cluster: cattle-system stuck Terminating, clusters that will not delete, the webhook that blocks everything, and getting the admin back

To wipe everything: the rancher-cleanup job and its verify job. For surgery: remove metadata finalizers (patch) and do not force the namespace spec.finalizers without first checking what resources remain; patch finalizers on clusters.provisioning and clusters.management; export and delete the rancher.cattle.io webhooks until the pod is back; reset-password inside the Rancher pod.