Nesta página, descrevemos como programar backups do Cassandra sem o Cloud Storage.
Neste método, os backups são armazenados em um servidor remoto especificado por você, em vez de um bucket do Cloud Storage. A Apigee usa
SSH para se comunicar com o servidor remoto.
É necessário programar os backups como jobs cron. Depois que uma programação de backup
é aplicada ao cluster híbrido, um job de backup do Kubernetes é
executado periodicamente de acordo com a programação no plano de ambiente de execução. O job aciona um script
de backup em cada nó do Cassandra no cluster híbrido que
coleta todos os dados no nó, cria um arquivo arquivado (compactado) e envia o arquivo ao servidor
especificado no arquivo overrides.yaml.
As etapas a seguir incluem exemplos comuns de como concluir tarefas específicas, como criar um par
de chaves SSH. Use os métodos apropriados para a instalação.
Selecionar um servidor de backup: escolha um servidor Linux ou Unix com armazenamento adequado para seus backups e verifique se ele pode ser acessado via SSH pelo plano de ambiente de execução híbrido da Apigee.
Configure o servidor SSH: instale um servidor SSH ou confirme se um servidor existente é seguro.
Criar um par de chaves SSH: gere um par de chaves SSH sem uma senha longa. Por exemplo:
Em que: exampleuser@example.com é uma string.
Qualquer string após -C no comando ssh-keygen se torna um comentário incluído na chave ssh recém-criada. A string de entrada pode ser qualquer string. Quando você usa um nome de conta no formato exampleuser@example.com, é possível identificar rapidamente qual conta corresponde à chave.
O comando vai gerar dois arquivos de chaves SSH, um de chave privada (por exemplo, "ssh_key.rsa") e um de chave pública (por exemplo, "ssh_key.pub").
Salve a chave privada em um local que o plano de ambiente de execução possa acessar.
Adicionar uma conta de usuário: no servidor de backup, crie um usuário chamado "apigee" com um diretório principal em /home/apigee. Verifique se
o novo usuário apigee tem um diretório principal em /home.
Configurar o diretório .ssh: no servidor de backup, crie um diretório .ssh em /home/apigee/.ssh. Exemplo:
cd /home/apigee
mkdir .sshcd .sshvi authorized_keys
Instalar a chave pública: coloque a chave pública no arquivo authorized_keys no diretório
/home/apigee/. O diretório de backup pode ser qualquer diretório, desde que o usuário apigee tenha acesso
a ele.Cole o conteúdo do arquivo ssh public key nele.
Verificar o acesso SSH: teste a conexão na máquina local ou em um nó do cluster:
O backup está desativado por padrão. É preciso definir essa
propriedade como true.
backup:keyFile
PATH_TO_PRIVATE_KEY_FILE
O caminho no sistema de arquivos local para o arquivo de chave privada SSH (chamado ssh_key
na etapa em que você criou o par de chaves SSH). O caminho precisa ser relativo ao diretório do gráfico apigee-datastore.
backup:server
BACKUP_SERVER_IP
O endereço IP do servidor de backup.
backup:storageDirectory
BACKUP_DIRECTORY
O nome do diretório de backup no servidor de backup. Precisa ser um diretório dentro de
home/apigee (o diretório de backup se chama cassandra_backup
na etapa em que você criou o diretório de backup).
backup:cloudProvider
GCP/HYBRID
Para um backup do Cloud Storage, defina a propriedade como GCP. Por exemplo, cloudProvider: "GCP".
Para um backup remoto do servidor, defina a propriedade como HYBRID. Por exemplo, cloudProvider: "HYBRID".
backup:schedule
SCHEDULE
O horário de início do backup, especificado na
sintaxe padrão de crontab. As horas estão no fuso horário local do cluster do Kubernetes. Padrão: 0 2 * * *
Aplique a configuração de backup ao escopo de armazenamento do
cluster:
Se você tiver problemas para acessar o servidor remoto no pod do Cassandra, verifique novamente a configuração do SSH no servidor remoto e verifique se a atualização do repositório de dados foi concluída.
Para verificar se o Cassandra usa a chave privada correta, execute o comando a seguir enquanto estiver conectado ao pod do Cassandra e compare a
saída com a chave privada que você criou:
[[["Fácil de entender","easyToUnderstand","thumb-up"],["Meu problema foi resolvido","solvedMyProblem","thumb-up"],["Outro","otherUp","thumb-up"]],[["Difícil de entender","hardToUnderstand","thumb-down"],["Informações incorretas ou exemplo de código","incorrectInformationOrSampleCode","thumb-down"],["Não contém as informações/amostras de que eu preciso","missingTheInformationSamplesINeed","thumb-down"],["Problema na tradução","translationIssue","thumb-down"],["Outro","otherDown","thumb-down"]],["Última atualização 2025-08-28 UTC."],[[["\u003cp\u003eThis guide explains how to configure scheduled backups for Cassandra data in Apigee Hybrid without using Cloud Storage, storing backups on a remote server accessible via SSH.\u003c/p\u003e\n"],["\u003cp\u003eTo set up the backup server, you must choose a Linux or Unix server with sufficient storage, configure an SSH server, create an SSH key pair without a passphrase, add an \u003ccode\u003eapigee\u003c/code\u003e user account, and place the SSH public key in the \u003ccode\u003eauthorized_keys\u003c/code\u003e file.\u003c/p\u003e\n"],["\u003cp\u003eBackup scheduling and destination are configured in the \u003ccode\u003eoverrides.yaml\u003c/code\u003e file by enabling backups, specifying the private key path, backup server IP, storage directory, setting \u003ccode\u003ecloudProvider\u003c/code\u003e to \u003ccode\u003eHYBRID\u003c/code\u003e, and defining a cron schedule.\u003c/p\u003e\n"],["\u003cp\u003eBackups are run as scheduled \u003ccode\u003ecron\u003c/code\u003e jobs, with a Kubernetes job periodically executing a backup script on each Cassandra node, archiving the data, and transferring it to the remote server, and you can also manually launch a backup if needed.\u003c/p\u003e\n"],["\u003cp\u003eYou should verify SSH access from the Cassandra pods to the backup server, ensuring no password or passphrase prompts, and confirm the correct private key is used by the Cassandra pods.\u003c/p\u003e\n"]]],[],null,["# Scheduling backups in a remote server\n\n| You are currently viewing version 1.12 of the Apigee hybrid documentation. **This version is end of life.** You should upgrade to a newer version. For more information, see [Supported versions](/apigee/docs/hybrid/supported-platforms#supported-versions).\n| **Warning:** Apigee Hybrid versions before 1.12.4 have [a known issue](/apigee/docs/release/known-issues#388608440) affecting backups that use `HYBRID` or `GCP` Cloud Providers. Please check the [Cleanup leftover snapshots](/apigee/docs/release/cassandra-ts) to check whether your setup is affected and for steps needed for resolution.\n\nThis page describes how to schedule backups for Cassandra without the Cloud Storage.\nIn this method, backups are stored on a remote server specified by you instead of a Cloud Storage bucket. Apigee uses\nSSH to communicate with the remote server.\n\nYou must schedule the backups as `cron` jobs. Once a backup schedule\nhas been applied to your hybrid cluster, a Kubernetes backup job is\nperiodically executed according to the schedule in the runtime plane. The job triggers a\nbackup script on each Cassandra node in your hybrid cluster that collects all the\ndata on the node, creates an archive (compressed) file of the data, and sends the archive\nto the server specified in your `overrides.yaml` file.\n| **Note:**\n|\n| - You must ensure there is enough space on the file system for the backups, and adjust the frequency of the backups to avoid unnecessarily filling the allotted storage space. Apigee does not dictate a retention policy for the backup files. You may want to create a retention policy for files appropriate to your installation.\n| - Applying backup configuration on the existing cluster will rolling restart Cassandra pods.\n\n\nThe following steps include common examples for completing specific tasks, like creating an SSH\nkey pair. Use the methods that are appropriate to your installation.\n\n\nThe procedure has the following parts:\n\n- [Set up the server and SSH](#server-ssh)\n- [Set the schedule and destination for backup](#overrides-backup)\n\n### Set up the server and SSH\n\n1. **Select a Backup Server:** Choose a Linux or Unix server with adequate storage for your backups and ensure it can be accessed via SSH from your Apigee hybrid runtime plane.\n2. **Configure the SSH Server:** Either install an SSH server or confirm that an existing one is secure.\n| **Caution:** For security purposes, make sure your SSH server is up to date.\n3. **Create an SSH Key Pair:** Generate an SSH key pair without a passphrase:For example: \n\n ssh-keygen -t rsa -b 4096 -C \u003cvar translate=\"no\"\u003e\u003cspan class=\"devsite-syntax-n\"\u003eexampleuser\u003c/span\u003e\u003cspan class=\"devsite-syntax-nv\"\u003e@example\u003c/span\u003e\u003cspan class=\"devsite-syntax-p\"\u003e.\u003c/span\u003e\u003cspan class=\"devsite-syntax-n\"\u003ecom\u003c/span\u003e\u003c/var\u003e\n Enter file in which to save the key (/Users/exampleuser/.ssh/id_rsa): $APIGEE_HOME/hybrid-files/certs/ssh_key\n Enter passphrase (empty for no passphrase):\n Enter same passphrase again:\n Your identification has been saved in ssh_key\n Your public key has been saved in ssh_key.pub\n The key fingerprint is:\n SHA256:DWKo334XMZcZYLOLrd/8HNpjTERPJJ0mc11UYmrPvSA exampleuser@example.com\n The key's randomart image is:\n +---[RSA 4096]----+\n | +. ++X|\n | . . o.=.*+|\n | . o . . o==o |\n | . . . =oo+o...|\n | . S +E oo .|\n | . . .. . o .|\n | . . . . o.. |\n | . ...o ++. |\n | .. .. +o+. |\n +----[SHA256]-----+\n\n\n Where: \u003cvar translate=\"no\"\u003eexampleuser@example.com\u003c/var\u003e is a string.\n Any string that follows `-C` in the `ssh-keygen`\n command becomes a comment included in the newly created `ssh`\n key. The input string can be any string. When you use an account name\n in the form of \u003cvar translate=\"no\"\u003eexampleuser@example.com\u003c/var\u003e, you can quickly\n identify which account goes with the key.\n\n The command will generate two SSH key files, A private key file (for example \\`ssh_key.rsa\\`) and a public key file (for example, \\`ssh_key.pub\\`).\n\n Save the private key to a location that your runtime plane can access.\n4. **Add a User Account:** On the backup server, create a user named apigee with a home directory under `/home/apigee`. Make sure the new `apigee` user has a home directory under `/home`.\n5. **Set Up the .ssh Directory:** On the backup server, create a `.ssh` directory in `/home/apigee/.ssh`. For example: \n\n cd /home/apigee\n mkdir .ssh\n cd .ssh\n vi authorized_keys\n\n6. **Install the Public Key:** Place the public key into the `authorized_keys` file within the `/home/apigee/` directory. The backup directory can be any directory as long as the `apigee` user has access to it.Paste the contents of the `ssh public key `file into the file.\n7. **Verify SSH Access:** Test the connection from your local machine or a cluster node: \n\n```\nssh -i PATH_TO_PRIVATE_KEY_FILE apigee@BACKUP_SERVER_IP\n```\n| **Note:** Make sure that you are able to access the remote server without being prompted for a password or a passphrase. If you get asked for a passphrase, please regenerate the ssh key pair without specifying a passphrase. In case you get prompted for a password, please double-check your ssh configuration on your remote server, since it is most likely not set up correctly.\n\n### Set the schedule and destination for backup\n\n\nYou set the schedule and destination for backups in your `overrides.yaml` file.\n\n1. Add the following parameters to your `overrides.yaml` file:\n\n ### Parameters\n\n ```actionscript-3\n cassandra:\n backup:\n enabled: true\n keyFile: \"\u003cvar translate=\"no\"\u003ePATH_TO_PRIVATE_KEY_FILE\u003c/var\u003e\"\n server: \"\u003cvar translate=\"no\"\u003eBACKUP_SERVER_IP\u003c/var\u003e\"\n storageDirectory: \"/home/apigee/\u003cvar translate=\"no\"\u003eBACKUP_DIRECTORY\u003c/var\u003e\"\n cloudProvider: \"HYBRID\" # required verbatim \"HYBRID\" (all caps)\n schedule: \"\u003cvar translate=\"no\"\u003eSCHEDULE\u003c/var\u003e\"\n ```\n\n ### Example\n\n ```actionscript-3\n cassandra:\n backup:\n enabled: true\n keyFile: \"private.key\" # path relative to apigee-datastore path\n server: \"34.56.78.90\"\n storageDirectory: \"/home/apigee/cassbackup\"\n cloudProvider: \"HYBRID\"\n schedule: \"0 2 * * *\"\n ```\n\n\n Where:\n\n2. Apply the backup configuration to the storage scope of your cluster: \n\n ```\n helm upgrade datastore apigee-datastore/ \\\n --install \\\n --namespace APIGEE_NAMESPACE \\\n --atomic \\\n -f OVERRIDES_FILE.yaml\n ```\n\n Where \u003cvar translate=\"no\"\u003eOVERRIDES_FILE\u003c/var\u003e is the path to the overrides file you just edited.\n3. Verify the backup job. For example: \n\n ```\n kubectl get cronjob -n APIGEE_NAMESPACE\n ``` \n\n ```component-pascal\n NAME SCHEDULE SUSPEND ACTIVE LAST SCHEDULE AGE\n apigee-cassandra-backup 33 * * * * False 0 \u003cnone\u003e 94s\n ```\n\n### Launch a manual backup\n\n\nBackup jobs are triggered automatically according to the cron schedule set in\n[cassandra.backup.schedule](/apigee/docs/hybrid/v1.12/config-prop-ref#cassandra-backup-schedule) in your\n`overrides.yaml` file. However, you can also initiate a backup job manually if needed\nusing the following command: \n\n```\nkubectl create job -n APIGEE_NAMESPACE --from=cronjob/apigee-cassandra-backup MANUAL_BACKUP_JOB_NAME\n```\n\nWhere \u003cvar translate=\"no\"\u003eMANUAL_BACKUP_JOB_NAME\u003c/var\u003e is the name of a manual backup job to be be created.\n\n### Troubleshooting\n\n1. Test the connection from a Cassandra pod. You need to make sure that your Cassandra pods can connect to your backup server using SSH:\n 1. Log into the shell of your Cassandra pod. For example: \n\n ```\n kubectl exec -it -n APIGEE_NAMESPACE APIGEE_CASSANDRA_POD -- /bin/bash\n ```\n\n\n Where \u003cvar translate=\"no\"\u003eAPIGEE_CASSANDRA_POD\u003c/var\u003e is the name of a Cassandra pod. Change this to\n the name of the pod you want to connect from.\n 2. Connect by SSH to your backup server, using the private SSH key mounted the Cassandra pod and server IP address: \n\n ```\n ssh -i /var/secrets/keys/key apigee@BACKUP_SERVER_IP\n ```\n | **Note:** You may see a warning at this point saying your server's fingerprint is unrecognized and asks if you would like to continue. You can continue at that prompt and verify the SSH configuration.\n2. If you have problems accessing your remote server from the Cassandra pod, please check your ssh configuration on the remote server again and also make sure that [upgrading the datastore](#overrides-backup) was successful.\n3. You can check if Cassandra uses the correct private key by running the following command while you are logged in to your Cassandra pod, and compare the output [with the private key you created](#create-key): \n\n ```\n cat /var/secrets/keys/key\n ```"]]
