navsection: installguide
title: Install Keepstore servers
...
+{% comment %}
+Copyright (C) The Arvados Authors. All rights reserved.
-We are going to install two Keepstore servers. By convention, we use the following hostname pattern:
+SPDX-License-Identifier: CC-BY-SA-3.0
+{% endcomment %}
+
+Keepstore provides access to underlying storage for reading and writing content-addressed blocks, with enforcement of Arvados permissions. Keepstore supports a variety of cloud object storage and POSIX filesystems for its backing store.
+
+We recommend starting off with two Keepstore servers. Exact server specifications will be site and workload specific, but in general keepstore will be I/O bound and should be set up to maximize aggregate bandwidth with compute nodes. To increase capacity (either space or throughput) it is straightforward to add additional servers, or (in cloud environments) to increase the machine size of the existing servers.
+
+By convention, we use the following hostname pattern:
<div class="offset1">
table(table table-bordered table-condensed).
|keep1.@uuid_prefix@.your.domain|
</div>
-Because the Keepstore servers are not directly accessible from the internet, these hostnames only need to resolve on the local network.
+Keepstore servers should not be directly accessible from the Internet (they are accessed via "keepproxy":install-keepproxy.html), so the hostnames only need to resolve on the private network.
h2. Install Keepstore
Verify that Keepstore is functional:
<notextile>
-<pre><code>~$ <span class="userinput">keepstore -h</span>
-2016/07/01 14:06:21 keepstore starting, pid 32339
-Usage of ./keepstore:
- -azure-max-get-bytes int
- Maximum bytes to request in a single GET request. If smaller than 67108864, use multiple concurrent range requests to retrieve a block. (default 67108864)
- -azure-storage-account-key-file string
- File containing the account key used for subsequent --azure-storage-container-volume arguments.
- -azure-storage-account-name string
- Azure storage account name used for subsequent --azure-storage-container-volume arguments.
- -azure-storage-container-volume value
- Use the given container as a storage volume. Can be given multiple times. (default [])
- -azure-storage-replication int
- Replication level to report to clients when data is stored in an Azure container. (default 3)
- -blob-signature-ttl int
- Lifetime of blob permission signatures in seconds. Modifying the ttl will invalidate all existing signatures. See services/api/config/application.default.yml. (default 1209600)
- -blob-signing-key-file string
- File containing the secret key for generating and verifying blob permission signatures.
- -data-manager-token-file string
- File with the API token used by the Data Manager. All DELETE requests or GET /index requests must carry this token.
- -enforce-permissions
- Enforce permission signatures on requests.
- -listen string
- Listening address, in the form "host:port". e.g., 10.0.1.24:8000. Omit the host part to listen on all interfaces. (default ":25107")
- -max-buffers int
- Maximum RAM to use for data buffers, given in multiples of block size (64 MiB). When this limit is reached, HTTP requests requiring buffers (like GET and PUT) will wait for buffer space to be released. (default 128)
- -max-requests int
- Maximum concurrent requests. When this limit is reached, new requests will receive 503 responses. Note: this limit does not include idle connections from clients using HTTP keepalive, so it does not strictly limit the number of concurrent connections. (default 2 * max-buffers)
- -never-delete
- If true, nothing will be deleted. Warning: the relevant features in keepstore and data manager have not been extensively tested. You should leave this option alone unless you can afford to lose data. (default true)
- -permission-key-file string
- Synonym for -blob-signing-key-file.
- -permission-ttl int
- Synonym for -blob-signature-ttl.
- -pid fuser -k pidfile
- Path to write pid file during startup. This file is kept open and locked with LOCK_EX until keepstore exits, so fuser -k pidfile is one way to shut down. Exit immediately if there is an error opening, locking, or writing the pid file.
- -readonly
- Do not write, delete, or touch anything on the following volumes.
- -s3-access-key-file string
- File containing the access key used for subsequent -s3-bucket-volume arguments.
- -s3-bucket-volume value
- Use the given bucket as a storage volume. Can be given multiple times. (default [])
- -s3-endpoint string
- Endpoint URL used for subsequent -s3-bucket-volume arguments. If blank, use the AWS endpoint corresponding to the -s3-region argument. For Google Storage, use "https://storage.googleapis.com".
- -s3-region string
- AWS region used for subsequent -s3-bucket-volume arguments. Allowed values are ["ap-southeast-1" "eu-west-1" "us-gov-west-1" "sa-east-1" "cn-north-1" "ap-northeast-1" "ap-southeast-2" "eu-central-1" "us-east-1" "us-west-1" "us-west-2"].
- -s3-replication int
- Replication level reported to clients for subsequent -s3-bucket-volume arguments. (default 2)
- -s3-secret-key-file string
- File containing the secret key used for subsequent -s3-bucket-volume arguments.
- -s3-unsafe-delete
- EXPERIMENTAL. Enable deletion (garbage collection), even though there are known race conditions that can cause data loss.
- -serialize
- Serialize read and write operations on the following volumes.
- -trash-check-interval duration
- Time duration at which the emptyTrash goroutine will check and delete expired trashed blocks. Default is one day. (default 24h0m0s)
- -trash-lifetime duration
- Time duration after a block is trashed during which it can be recovered using an /untrash request
- -volume value
- Local storage directory. Can be given more than once to add multiple directories. If none are supplied, the default is to use all directories named "keep" that exist in the top level directory of a mount point at startup time. Can be a comma-separated list, but this is deprecated: use multiple -volume arguments instead. (default [])
- -volumes value
- Deprecated synonym for -volume. (default [])
+<pre><code>~$ <span class="userinput">keepstore --version</span>
</code></pre>
</notextile>
-h3. Prepare storage volumes
-
-{% include 'notebox_begin' %}
-This section uses a local filesystem as a backing store. If you are using Azure Storage, follow the setup instructions on the "Azure Blob Storage":configure-azure-blob-storage.html page instead.
-{% include 'notebox_end' %}
-
-There are two ways to specify a set of local directories where keepstore should store its data files.
-# Implicitly, by creating a directory called @keep@ at the top level of each filesystem you intend to use, and omitting @-volume@ arguments.
-# Explicitly, by providing a @-volume@ argument for each directory.
-
-For example, if there are filesystems mounted at @/mnt@ and @/mnt2@:
-
-<notextile>
-<pre><code>~$ <span class="userinput">mkdir /mnt/keep /mnt2/keep</span>
-~$ <span class="userinput">keepstore</span>
-2015/05/08 13:44:26 keepstore starting, pid 2765
-2015/05/08 13:44:26 Using volume [UnixVolume /mnt/keep] (writable=true)
-2015/05/08 13:44:26 Using volume [UnixVolume /mnt2/keep] (writable=true)
-2015/05/08 13:44:26 listening at :25107
-</code></pre>
-</notextile>
-
-Equivalently:
-
-<notextile>
-<pre><code>~$ <span class="userinput">mkdir /mnt/keep /mnt2/keep</span>
-~$ <span class="userinput">keepstore -volume=/mnt/keep -volume=/mnt2/keep</span>
-2015/05/08 13:44:26 keepstore starting, pid 2765
-2015/05/08 13:44:26 Using volume [UnixVolume /mnt/keep] (writable=true)
-2015/05/08 13:44:26 Using volume [UnixVolume /mnt2/keep] (writable=true)
-2015/05/08 13:44:26 listening at :25107
-</code></pre>
-</notextile>
+h3. Create config file
+
+By default, keepstore will look for its configuration file at @/etc/arvados/keepstore/keepstore.yml@
+
+You can override the configuration file location using the @-config@ command line option to keepstore.
+
+The following is a sample configuration file:
+
+<pre>
+# Duration for which new permission signatures (returned in PUT
+# responses) will be valid. This should be equal to the API
+# server's blob_signature_ttl configuration entry.
+BlobSignatureTTL: 336h0m0s
+
+# Local file containing the secret blob signing key (used to generate
+# and verify blob signatures). The contents of the key file must be
+# identical to the API server's blob_signing_key configuration entry.
+BlobSigningKeyFile: ""
+
+# Print extra debug logging
+Debug: false
+
+# Maximum number of concurrent block deletion operations (per
+# volume) when emptying trash. Default is 1.
+EmptyTrashWorkers: 1
+
+# Enable trash and delete features. If false, trash lists will be
+# accepted but blocks will not be trashed or deleted.
+# Keepstore does not delete data on its own. The keep-balance
+# service determines which blocks are candidates for deletion
+# and instructs the keepstore to move those blocks to the trash.
+EnableDelete: true
+
+# Local port to listen on. Can be 'address:port' or ':port', where
+# 'address' is a host IP address or name and 'port' is a port number
+# or name.
+Listen: :25107
+
+# Format of request/response and error logs: "json" or "text".
+LogFormat: json
+
+# The secret key that must be provided by monitoring services
+# wishing to access the health check endpoint (/_health).
+ManagementToken: ""
+
+# Maximum RAM to use for data buffers, given in multiples of block
+# size (64 MiB). When this limit is reached, HTTP requests requiring
+# buffers (like GET and PUT) will wait for buffer space to be
+# released.
+#
+# It should be set such that MaxBuffers * 64MiB + 10% fits
+# comfortably in memory. On a host dedicated to running keepstore,
+# divide total memory by 88MiB to suggest a suitable value. For example,
+# if grep MemTotal /proc/meminfo reports MemTotal: 7125440 kB,
+# compute 7125440 / (88 * 1024)=79 and configure MaxBuffers: 79
+MaxBuffers: 128
+
+# Maximum concurrent requests. When this limit is reached, new
+# requests will receive 503 responses. Note: this limit does not
+# include idle connections from clients using HTTP keepalive, so it
+# does not strictly limit the number of concurrent connections. If
+# omitted or zero, the default is 2 * MaxBuffers.
+MaxRequests: 0
+
+# Path to write PID file during startup. This file is kept open and
+# locked with LOCK_EX until keepstore exits, so "fuser -k pidfile" is
+# one way to shut down. Exit immediately if there is an error
+# opening, locking, or writing the PID file.
+PIDFile: ""
+
+# Maximum number of concurrent pull operations. Default is 1, i.e.,
+# pull lists are processed serially. A pull operation copies a block
+# from another keepstore server.
+PullWorkers: 1
+
+# Honor read requests only if a valid signature is provided. This
+# should be true, except for development use and when migrating from
+# a very old version.
+RequireSignatures: true
+
+# Local file containing the Arvados API token used by keep-balance
+# or data manager. Delete, trash, and index requests are honored
+# only for this token.
+SystemAuthTokenFile: ""
+
+# Path to server certificate file in X509 format. Enables TLS mode.
+#
+# Example: /var/lib/acme/live/keep0.example.com/fullchain
+TLSCertificateFile: ""
+
+# Path to server key file in X509 format. Enables TLS mode.
+#
+# The key pair is read from disk during startup, and whenever SIGHUP
+# is received.
+#
+# Example: /var/lib/acme/live/keep0.example.com/privkey
+TLSKeyFile: ""
+
+# How often to check for (and delete) trashed blocks whose
+# TrashLifetime has expired.
+TrashCheckInterval: 24h0m0s
+
+# Time duration after a block is trashed during which it can be
+# recovered using an /untrash request.
+TrashLifetime: 336h0m0s
+
+# Maximum number of concurrent trash operations (moving a block to the
+# trash, or permanently deleting it) . Default is 1, i.e., trash lists
+# are processed serially. If individual trash operations have high
+# latency (eg some cloud platforms) you should increase this.
+TrashWorkers: 1
+</pre>
+
+h3. Notes on storage management
+
+On its own, a keepstore server never deletes data. The "keep-balance":install-keep-balance.html service determines which blocks are candidates for deletion and instructs the keepstore to move those blocks to the trash.
+
+When a block is newly written, it is protected from deletion for the duration in @BlobSignatureTTL@. During this time, it cannot be trashed.
+
+If keep-balance instructs keepstore to trash a block which is older than @BlobSignatureTTL@, and @EnableDelete@ is true, the block will be moved to "trash". A block which is in the trash is no longer accessible by read requests, but has not yet been permanently deleted. Blocks which are in the trash may be recovered using the "untrash" API endpoint. Blocks are permanently deleted after they have been in the trash for the duration in @TrashLifetime@.
+
+Keep-balance is also responsible for balancing the distribution of blocks across keepstore servers by asking servers to pull blocks from other servers (as determined by their "storage class":{{site.baseurl}}/admin/storage-classes.html and "rendezvous hashing order":{{site.baseurl}}/api/storage.html). Pulling a block makes a copy. If a block is overreplicated (i.e. there are excess copies) after pulling, it will be subsequently trashed on the original server.
+
+h3. Configure storage volumes
+
+Available storage volume types include POSIX filesystems and cloud object storage.
+
+* To use a POSIX filesystem, including both local filesystems (ext4, xfs) and network file system such as GPFS or Lustre, follow the setup instructions on "Filesystem storage":configure-fs-storage.html
+* If you are using S3-compatible object storage (including Amazon S3, Google Cloud Storage, and Ceph RADOS), follow the setup instructions on "S3 Object Storage":configure-s3-object-storage.html
+* If you are using Azure Blob Storage, follow the setup instructions on "Azure Blob Storage":configure-azure-blob-storage.html
h3. Run keepstore as a supervised service
Install runit to supervise the keepstore daemon. {% include 'install_runit' %}
-Install this script as the run script for the keepstore service, modifying it as directed below.
+Install this script as the run script @/etc/sv/keepstore/run@ for the keepstore service:
<notextile>
<pre><code>#!/bin/sh
exec 2>&1
-exec GOGC=10 GOMAXPROCS=<span class="userinput">4</span> keepstore \
- -enforce-permissions=true \
- -blob-signing-key-file=<span class="userinput">/etc/keepstore/blob-signing.key</span> \
- -max-buffers=<span class="userinput">100</span> \
- -serialize=true \
- -never-delete=false \
- -volume=<span class="userinput">/mnt/keep</span> \
- -volume=<span class="userinput">/mnt2/keep</span>
+GOGC=10 exec keepstore -config /etc/arvados/keepstore/keepstore.yml
</code></pre>
</notextile>
-The @GOMAXPROCS@ environment variable determines the maximum number of concurrent threads, and should normally be set to the number of CPU cores present.
+h3. Set up additional servers
-The @-max-buffers@ argument limits keepstore's memory usage. It should be set such that @max-buffers * 64MiB + 10%@ fits comfortably in memory. For example, @-max-buffers=100@ is suitable for a host with 8 GiB RAM.
+Repeat the above sections to prepare volumes and bring up supervised services on each Keepstore server you are setting up.
-If you want access control on your Keepstore server(s), you must specify the @-enforce-permissions@ flag and provide a signing key. The @-blob-signing-key-file@ argument should be a file containing a long random alphanumeric string with no internal line breaks (it is also possible to use a socket or FIFO: keepstore reads it only once, at startup). This key must be the same as the @blob_signing_key@ configured in the "API server's":install-api-server.html configuration file, @/etc/arvados/api/application.yml@.
+h3. Tell the API server about the Keepstore servers
-The @-serialize=true@ (default: @false@) argument limits keepstore to one reader/writer process per storage partition. This avoids thrashing by allowing the storage device underneath the storage partition to do read/write operations sequentially. Enabling @-serialize@ can improve Keepstore performance if the storage partitions map 1:1 to physical disks that are dedicated to Keepstore, particularly so for mechanical disks. In some cloud environments, enabling @-serialize@ has also also proven to be beneficial for performance, but YMMV. If your storage partition(s) are backed by network or RAID storage that can handle many simultaneous reader/writer processes without thrashing, you probably do not want to set @-serialize@.
+The API server needs to be informed about the presence of your Keepstore servers.
-h3. Set up additional servers
+First, if you don't already have an admin token, create a superuser token:
-Repeat the above sections to prepare volumes and bring up supervised services on each Keepstore server you are setting up.
+{% include 'create_superuser_token' %}
-h3. Tell the API server about the Keepstore servers
+Configure your environment to run @arv@ using the output of create_superuser_token.rb:
-The API server needs to be informed about the presence of your Keepstore servers. For each of the Keepstore servers you have created, please execute the following commands on your <strong>shell server</strong>.
+<pre>
+export ARVADOS_API_HOST=zzzzz.example.com
+export ARVADOS_API_TOKEN=zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
+</pre>
-Make sure to update the @service_host@ value to match each of your Keepstore servers.
+Use this command to register each keepstore server you have installed. Make sure to update the @service_host@ value.
<notextile>
<pre><code>~$ <span class="userinput">prefix=`arv --format=uuid user current | cut -d- -f1`</span>
~$ <span class="userinput">echo "Site prefix is '$prefix'"</span>
~$ <span class="userinput">read -rd $'\000' keepservice <<EOF; arv keep_service create --keep-service "$keepservice"</span>
<span class="userinput">{
- "service_host":"<strong>keep0.$prefix.your.domain</strong>",
+ "service_host":"<strong>keep0.$uuid_prefix.your.domain</strong>",
"service_port":25107,
"service_ssl_flag":false,
"service_type":"disk"
}
EOF</span>
</code></pre></notextile>
+
+h3(#testing). Testing keep
+
+Install the "Python SDK":{{site.baseurl}}/sdk/python/sdk-python.html
+
+@ARVADOS_API_HOST@ and @ARVADOS_API_TOKEN@ must be set in the environment.
+
+You should now be able to use @arv-put@ to upload collections and @arv-get@ to fetch collections:
+
+<pre>
+$ echo "hello world!" > hello.txt
+
+$ arv-put --portable-data-hash hello.txt
+2018-07-12 13:35:25 arvados.arv_put[28702] INFO: Creating new cache file at /home/example/.cache/arvados/arv-put/1571ec0adb397c6a18d5c74cc95b3a2a
+0M / 0M 100.0% 2018-07-12 13:35:27 arvados.arv_put[28702] INFO:
+
+2018-07-12 13:35:27 arvados.arv_put[28702] INFO: Collection saved as 'Saved at 2018-07-12 17:35:25 UTC by example@example'
+59389a8f9ee9d399be35462a0f92541c+53
+
+$ arv-get 59389a8f9ee9d399be35462a0f92541c+53/hello.txt
+hello world!
+</pre>
projects / arvados.git / blobdiff