From 3ffc8ac5fa560a87ae7c4a3a0e37957c26926729 Mon Sep 17 00:00:00 2001 From: Sergey Dudoladov Date: Fri, 31 May 2019 15:51:58 +0200 Subject: [PATCH] Sample logical backup image. (#572) * Sample logical backup image. Based on the earlier work by Dmitry Dolgov @erthalion --- docker/logical-backup/Dockerfile | 33 ++++++++++ docker/logical-backup/dump.sh | 94 +++++++++++++++++++++++++++ docs/administrator.md | 2 +- docs/reference/operator_parameters.md | 4 +- 4 files changed, 130 insertions(+), 3 deletions(-) create mode 100644 docker/logical-backup/Dockerfile create mode 100755 docker/logical-backup/dump.sh diff --git a/docker/logical-backup/Dockerfile b/docker/logical-backup/Dockerfile new file mode 100644 index 000000000..1da6f7386 --- /dev/null +++ b/docker/logical-backup/Dockerfile @@ -0,0 +1,33 @@ +FROM ubuntu:18.04 +LABEL maintainer="Team ACID @ Zalando " + +SHELL ["/bin/bash", "-o", "pipefail", "-c"] +RUN apt-get update \ + && apt-get install --no-install-recommends -y \ + apt-utils \ + ca-certificates \ + lsb-release \ + pigz \ + python3-pip \ + python3-setuptools \ + curl \ + jq \ + gnupg \ + && pip3 install --no-cache-dir awscli --upgrade \ + && echo "deb http://apt.postgresql.org/pub/repos/apt/ $(lsb_release -cs)-pgdg main" > /etc/apt/sources.list.d/pgdg.list \ + && cat /etc/apt/sources.list.d/pgdg.list \ + && curl --silent https://www.postgresql.org/media/keys/ACCC4CF8.asc | apt-key add - \ + && apt-get update \ + && apt-get install --no-install-recommends -y \ + postgresql-client-11 \ + postgresql-client-10 \ + postgresql-client-9.6 \ + postgresql-client-9.5 \ + && apt-get clean \ + && rm -rf /var/lib/apt/lists/* + +COPY dump.sh ./ + +ENV PG_DIR=/usr/lib/postgresql/ + +ENTRYPOINT ["/dump.sh"] diff --git a/docker/logical-backup/dump.sh b/docker/logical-backup/dump.sh new file mode 100755 index 000000000..d74c7653b --- /dev/null +++ b/docker/logical-backup/dump.sh @@ -0,0 +1,94 @@ +#! /usr/bin/env bash + +# enable unofficial bash strict mode +set -o errexit +set -o nounset +set -o pipefail +IFS=$'\n\t' + +# make script trace visible via `kubectl logs` +set -o xtrace + +ALL_DB_SIZE_QUERY="select sum(pg_database_size(datname)::numeric) from pg_database;" +PG_BIN=$PG_DIR/$PG_VERSION/bin +DUMP_SIZE_COEFF=5 + +TOKEN=$(cat /var/run/secrets/kubernetes.io/serviceaccount/token) +K8S_API_URL=https://$KUBERNETES_SERVICE_HOST:$KUBERNETES_SERVICE_PORT/api/v1 +CERT=/var/run/secrets/kubernetes.io/serviceaccount/ca.crt + +function estimate_size { + "$PG_BIN"/psql -tqAc "${ALL_DB_SIZE_QUERY}" +} + +function dump { + # settings are taken from the environment + "$PG_BIN"/pg_dumpall +} + +function compress { + pigz +} + +function aws_upload { + declare -r EXPECTED_SIZE="$1" + + # mimic bucket setup from Spilo + # to keep logical backups at the same path as WAL + # NB: $LOGICAL_BACKUP_S3_BUCKET_SCOPE_SUFFIX already contains the leading "/" when set by the Postgres operator + PATH_TO_BACKUP=s3://$LOGICAL_BACKUP_S3_BUCKET"/spilo/"$SCOPE$LOGICAL_BACKUP_S3_BUCKET_SCOPE_SUFFIX"/logical_backups/"$(date +%s).sql.gz + + if [ -z "$EXPECTED_SIZE" ]; then + aws s3 cp - "$PATH_TO_BACKUP" --debug --sse="AES256" + else + aws s3 cp - "$PATH_TO_BACKUP" --debug --expected-size "$EXPECTED_SIZE" --sse="AES256" + fi; +} + +function get_pods { + declare -r SELECTOR="$1" + + curl "${K8S_API_URL}/pods?$SELECTOR" \ + --cacert $CERT \ + -H "Authorization: Bearer ${TOKEN}" | jq .items[].status.podIP -r +} + +function get_current_pod { + curl "${K8S_API_URL}/pods?fieldSelector=metadata.name%3D${HOSTNAME}" \ + --cacert $CERT \ + -H "Authorization: Bearer ${TOKEN}" +} + +declare -a search_strategy=( + list_all_replica_pods_current_node + list_all_replica_pods_any_node + get_master_pod +) + +function list_all_replica_pods_current_node { + get_pods "labelSelector=version%3D${SCOPE},spilo-role%3Dreplica&fieldSelector=spec.nodeName%3D${CURRENT_NODENAME}" | head -n 1 +} + +function list_all_replica_pods_any_node { + get_pods "labelSelector=version%3D${SCOPE},spilo-role%3Dreplica" | head -n 1 +} + +function get_master_pod { + get_pods "labelSelector=version%3D${SCOPE},spilo-role%3Dmaster" | head -n 1 +} + +CURRENT_NODENAME=$(get_current_pod | jq .items[].spec.nodeName --raw-output) +export CURRENT_NODENAME + +for search in "${search_strategy[@]}"; do + + PGHOST=$(eval "$search") + export PGHOST + + if [ -n "$PGHOST" ]; then + break + fi + +done + +dump | compress | aws_upload $(($(estimate_size) / DUMP_SIZE_COEFF)) diff --git a/docs/administrator.md b/docs/administrator.md index f6f37aafb..ab8a51877 100644 --- a/docs/administrator.md +++ b/docs/administrator.md @@ -346,7 +346,7 @@ The operator logs reasons for a rolling update with the `info` level and a diff The operator can manage k8s cron jobs to run logical backups of Postgres clusters. The cron job periodically spawns a batch job that runs a single pod. The backup script within this pod's container can connect to a DB for a logical backup. The operator updates cron jobs during Sync if the job schedule changes; the job name acts as the job identifier. These jobs are to be enabled for each indvidual Postgres cluster by setting `enableLogicalBackup: true` in its manifest. Notes: -1. The provided `registry.opensource.zalan.do/acid/logical-backup` image implements the backup via `pg_dumpall` and upload of (compressed) results to an S3 bucket; `pg_dumpall` requires a `superuser` access to a DB and runs on the replica when possible. +1. The [example image](../docker/logical-backup/Dockerfile) implements the backup via `pg_dumpall` and upload of compressed and encrypted results to an S3 bucket; the default image ``registry.opensource.zalan.do/acid/logical-backup`` is the same image built with the Zalando-internal CI pipeline. `pg_dumpall` requires a `superuser` access to a DB and runs on the replica when possible. 2. Due to the [limitation of Kubernetes cron jobs](https://kubernetes.io/docs/concepts/workloads/controllers/cron-jobs/#cron-job-limitations) it is highly advisable to set up additional monitoring for this feature; such monitoring is outside of the scope of operator responsibilities. diff --git a/docs/reference/operator_parameters.md b/docs/reference/operator_parameters.md index 3e06cf31d..592fef5b8 100644 --- a/docs/reference/operator_parameters.md +++ b/docs/reference/operator_parameters.md @@ -485,9 +485,9 @@ scalyr sidecar. In the CRD-based configuration they are grouped under the * **logical_backup_schedule** Backup schedule in the cron format. Please take [the reference schedule format](https://kubernetes.io/docs/tasks/job/automated-tasks-with-cron-jobs/#schedule) into account. Default: "30 00 \* \* \*" - + * **logical_backup_docker_image** - Docker image for the pods of the cron job. Must implement backup logic and correctly handle pod and job restarts. The default image runs `pg_dumpall` (on a replica if possible) and uploads compressed results to an S3 bucket under the key `/spilo/pg_cluster_name/cluster_k8s_uuid/logical_backups` Default: "registry.opensource.zalan.do/acid/logical-backup" + An image for pods of the logical backup job. The [example image](../../docker/logical-backup/Dockerfile) runs `pg_dumpall` on a replica if possible and uploads compressed results to an S3 bucket under the key `/spilo/pg_cluster_name/cluster_k8s_uuid/logical_backups`. The default image is the same image built with the Zalando-internal CI pipeline. Default: "registry.opensource.zalan.do/acid/logical-backup" * **logical_backup_s3_bucket** S3 bucket to store backup results. The bucket has to be present and accessible by Postgres pods. Default: empty.