* Stop specifying logbufs=8; that's the default * Stop including nodiratime with noatime; the latter implies the former Nothing wrong with being explicit, I suppose, but may as well keep the mount options to what we can easily explain: we want noatime because Swift does not use atime, so we don't want to lose any performance to tracking atime. Change-Id: I1e52b4368ad7eb375964eee5132bc50297536355
26 KiB
SAIO (Swift All In One)
Note
This guide assumes an existing Linux server. A physical machine or VM will work. We recommend configuring it with at least 2GB of memory and 40GB of storage space. We recommend using a VM in order to isolate Swift and its dependencies from other projects you may be working on.
Instructions for setting up a development VM
This section documents setting up a virtual machine for doing Swift development. The virtual machine will emulate running a four node Swift cluster. To begin:
- Get a Linux system server image, this guide will cover:
- Ubuntu 14.04, 16.04 LTS
- CentOS 7
- Fedora
- OpenSuse
- Create guest virtual machine from the image.
What's in a <your-user-name>
Much of the configuration described in this guide requires escalated
administrator (root) privileges; however, we assume that
administrator logs in as an unprivileged user and can use
sudo to run privileged commands.
Swift processes also run under a separate user and group, set by
configuration option, and referenced as
<your-user-name>:<your-group-name>. The default
user is swift, which may not exist on your system. These
instructions are intended to allow a developer to use his/her username
for <your-user-name>:<your-group-name>.
Note
For OpenSuse users, a user's primary group is users, so
you have 2 options:
Change
${USER}:${USER}to${USER}:usersin all references of this guide; orCreate a group for your username and add yourself to it:
sudo groupadd ${USER} && sudo gpasswd -a ${USER} ${USER} && newgrp ${USER}
Installing dependencies
On
aptbased systems:sudo apt-get update sudo apt-get install curl gcc memcached rsync sqlite3 xfsprogs \ git-core libffi-dev python-setuptools \ liberasurecode-dev libssl-dev sudo apt-get install python-coverage python-dev python-nose \ python-xattr python-eventlet \ python-greenlet python-pastedeploy \ python-netifaces python-pip python-dnspython \ python-mockOn
CentOS(requires additional repositories):sudo yum update sudo yum install epel-release sudo yum-config-manager --enable epel extras sudo yum install centos-release-openstack-train sudo yum install curl gcc memcached rsync sqlite xfsprogs git-core \ libffi-devel xinetd liberasurecode-devel \ openssl-devel python-setuptools \ python-coverage python-devel python-nose \ pyxattr python-eventlet \ python-greenlet python-paste-deploy \ python-netifaces python-pip python-dns \ python-mockOn
Fedora:sudo dnf update sudo dnf install curl gcc memcached rsync-daemon sqlite xfsprogs git-core \ libffi-devel xinetd liberasurecode-devel \ openssl-devel python-setuptools \ python-coverage python-devel python-nose \ pyxattr python-eventlet \ python-greenlet python-paste-deploy \ python-netifaces python-pip python-dns \ python-mockOn
OpenSuse:sudo zypper install curl gcc memcached rsync sqlite3 xfsprogs git-core \ libffi-devel liberasurecode-devel python2-setuptools \ libopenssl-devel sudo zypper install python2-coverage python-devel python2-nose \ python-xattr python-eventlet python2-greenlet \ python2-netifaces python2-pip python2-dnspython \ python2-mock
Note
This installs necessary system dependencies and most of the python dependencies. Later in the process setuptools/distribute or pip will install and/or upgrade packages.
Configuring storage
Swift requires some space on XFS filesystems to store data and run tests.
Choose either partition-section or loopback-section.
Using a partition for storage
If you are going to use a separate partition for Swift data, be sure to add another device when creating the VM, and follow these instructions:
Note
The disk does not have to be /dev/sdb1 (for example, it
could be /dev/vdb1) however the mount point should still be
/mnt/sdb1.
Set up a single partition on the device (this will wipe the drive):
sudo parted /dev/sdb mklabel msdos mkpart p xfs 0% 100%Create an XFS file system on the partition:
sudo mkfs.xfs /dev/sdb1Find the UUID of the new partition:
sudo blkidEdit
/etc/fstaband add:UUID="<UUID-from-output-above>" /mnt/sdb1 xfs noatime 0 0Create the Swift data mount point and test that mounting works:
sudo mkdir /mnt/sdb1 sudo mount -aNext, skip to
common-dev-section.
Using a loopback device for storage
If you want to use a loopback device instead of another partition, follow these instructions:
Create the file for the loopback device:
sudo mkdir -p /srv sudo truncate -s 1GB /srv/swift-disk sudo mkfs.xfs /srv/swift-diskModify size specified in the
truncatecommand to make a larger or smaller partition as needed.Edit /etc/fstab and add:
/srv/swift-disk /mnt/sdb1 xfs loop,noatime 0 0Create the Swift data mount point and test that mounting works:
sudo mkdir /mnt/sdb1 sudo mount -a
Common Post-Device Setup
Create the individualized data links:
sudo mkdir /mnt/sdb1/1 /mnt/sdb1/2 /mnt/sdb1/3 /mnt/sdb1/4 sudo chown ${USER}:${USER} /mnt/sdb1/* for x in {1..4}; do sudo ln -s /mnt/sdb1/$x /srv/$x; done sudo mkdir -p /srv/1/node/sdb1 /srv/1/node/sdb5 \ /srv/2/node/sdb2 /srv/2/node/sdb6 \ /srv/3/node/sdb3 /srv/3/node/sdb7 \ /srv/4/node/sdb4 /srv/4/node/sdb8 sudo mkdir -p /var/run/swift sudo mkdir -p /var/cache/swift /var/cache/swift2 \ /var/cache/swift3 /var/cache/swift4 sudo chown -R ${USER}:${USER} /var/run/swift sudo chown -R ${USER}:${USER} /var/cache/swift* # **Make sure to include the trailing slash after /srv/$x/** for x in {1..4}; do sudo chown -R ${USER}:${USER} /srv/$x/; doneNote
We create the mount points and mount the loopback file under /mnt/sdb1. This file will contain one directory per simulated Swift node, each owned by the current Swift user.
We then create symlinks to these directories under /srv. If the disk sdb or loopback file is unmounted, files will not be written under /srv/*, because the symbolic link destination /mnt/sdb1/* will not exist. This prevents disk sync operations from writing to the root partition in the event a drive is unmounted.
Restore appropriate permissions on reboot.
On traditional Linux systems, add the following lines to
/etc/rc.local(before theexit 0):mkdir -p /var/cache/swift /var/cache/swift2 /var/cache/swift3 /var/cache/swift4 chown <your-user-name>:<your-group-name> /var/cache/swift* mkdir -p /var/run/swift chown <your-user-name>:<your-group-name> /var/run/swiftOn CentOS and Fedora we can use systemd (rc.local is deprecated):
cat << EOF |sudo tee /etc/tmpfiles.d/swift.conf d /var/cache/swift 0755 ${USER} ${USER} - - d /var/cache/swift2 0755 ${USER} ${USER} - - d /var/cache/swift3 0755 ${USER} ${USER} - - d /var/cache/swift4 0755 ${USER} ${USER} - - d /var/run/swift 0755 ${USER} ${USER} - - EOFOn OpenSuse place the lines in
/etc/init.d/boot.local.
Note
On some systems the rc file might need to be an executable shell script.
Creating an XFS tmp dir
Tests require having a directory available on an XFS filesystem. By
default the tests use /tmp, however this can be pointed
elsewhere with the TMPDIR environment variable.
Note
If your root filesystem is XFS, you can skip this section if
/tmp is just a directory and not a mounted tmpfs. Or you
could simply point to any existing directory owned by your user by
specifying it with the TMPDIR environment variable.
If your root filesystem is not XFS, you should create a loopback
device, format it with XFS and mount it. You can mount it over
/tmp or to another location and specify it with the
TMPDIR environment variable.
Create the file for the tmp loopback device:
sudo mkdir -p /srv sudo truncate -s 1GB /srv/swift-tmp # create 1GB file for XFS in /srv sudo mkfs.xfs /srv/swift-tmpTo mount the tmp loopback device at
/tmp, do the following:sudo mount -o loop,noatime /srv/swift-tmp /tmp sudo chmod -R 1777 /tmpTo persist this, edit and add the following to
/etc/fstab:/srv/swift-tmp /tmp xfs rw,noatime,attr2,inode64,noquota 0 0
To mount the tmp loopback at an alternate location (for example,
/mnt/tmp), do the following:sudo mkdir -p /mnt/tmp sudo mount -o loop,noatime /srv/swift-tmp /mnt/tmp sudo chown ${USER}:${USER} /mnt/tmpTo persist this, edit and add the following to
/etc/fstab:/srv/swift-tmp /mnt/tmp xfs rw,noatime,attr2,inode64,noquota 0 0Set your
TMPDIRenvironment dir so that Swift looks in the right location:export TMPDIR=/mnt/tmp echo "export TMPDIR=/mnt/tmp" >> $HOME/.bashrc
Getting the code
Check out the python-swiftclient repo:
cd $HOME; git clone https://github.com/openstack/python-swiftclient.gitBuild a development installation of python-swiftclient:
cd $HOME/python-swiftclient; sudo python setup.py develop; cd -Ubuntu 12.04 users need to install python-swiftclient's dependencies before the installation of python-swiftclient. This is due to a bug in an older version of setup tools:
cd $HOME/python-swiftclient; sudo pip install -r requirements.txt; sudo python setup.py develop; cd -Check out the Swift repo:
git clone https://github.com/openstack/swift.gitBuild a development installation of Swift:
cd $HOME/swift; sudo pip install --no-binary cryptography -r requirements.txt; sudo python setup.py develop; cd -Note
Due to a difference in how
libssl.sois named in OpenSuse vs. other Linux distros the wheel/binary won't work; thus we use--no-binary cryptographyto buildcryptographylocally.Fedora users might have to perform the following if development installation of Swift fails:
sudo pip install -U xattrInstall Swift's test dependencies:
cd $HOME/swift; sudo pip install -r test-requirements.txt
Setting up rsync
Create
/etc/rsyncd.conf:sudo cp $HOME/swift/doc/saio/rsyncd.conf /etc/ sudo sed -i "s/<your-user-name>/${USER}/" /etc/rsyncd.confHere is the default
rsyncd.conffile contents maintained in the repo that is copied and fixed up above:/../saio/rsyncd.conf
Enable rsync daemon
On Ubuntu, edit the following line in
/etc/default/rsync:RSYNC_ENABLE=true
Note
You might have to create the file to perform the edits.
On CentOS and Fedora, enable the systemd service:
sudo systemctl enable rsyncdOn OpenSuse, nothing needs to happen here.
On platforms with SELinux in
Enforcingmode, either set toPermissive:sudo setenforce Permissive sudo sed -i 's/^SELINUX=.*/SELINUX=permissive/g' /etc/selinux/configOr just allow rsync full access:
sudo setsebool -P rsync_full_access 1Start the rsync daemon
On Ubuntu 14.04, run:
sudo service rsync restartOn Ubuntu 16.04, run:
sudo systemctl enable rsync sudo systemctl start rsyncOn CentOS, Fedora and OpenSuse, run:
sudo systemctl start rsyncdOn other xinetd based systems simply run:
sudo service xinetd restart
Verify rsync is accepting connections for all servers:
rsync rsync://pub@localhost/You should see the following output from the above command:
account6012 account6022 account6032 account6042 container6011 container6021 container6031 container6041 object6010 object6020 object6030 object6040
Starting memcached
On non-Ubuntu distros you need to ensure memcached is running:
sudo service memcached start
sudo chkconfig memcached on
or:
sudo systemctl enable memcached
sudo systemctl start memcached
The tempauth middleware stores tokens in memcached. If memcached is not running, tokens cannot be validated, and accessing Swift becomes impossible.
Optional: Setting up rsyslog for individual logging
Fedora and OpenSuse may not have rsyslog installed, in which case you will need to install it if you want to use individual logging.
Install rsyslogd
On Fedora:
sudo dnf install rsyslogOn OpenSuse:
sudo zypper install rsyslog
Install the Swift rsyslogd configuration:
sudo cp $HOME/swift/doc/saio/rsyslog.d/10-swift.conf /etc/rsyslog.d/Be sure to review that conf file to determine if you want all the logs in one file vs. all the logs separated out, and if you want hourly logs for stats processing. For convenience, we provide its default contents below:
/../saio/rsyslog.d/10-swift.conf
Edit
/etc/rsyslog.confand make the following change (usually in the "GLOBAL DIRECTIVES" section):$PrivDropToGroup admIf using hourly logs (see above) perform:
sudo mkdir -p /var/log/swift/hourlyOtherwise perform:
sudo mkdir -p /var/log/swiftSetup the logging directory and start syslog:
On Ubuntu:
sudo chown -R syslog.adm /var/log/swift sudo chmod -R g+w /var/log/swift sudo service rsyslog restartOn CentOS, Fedora and OpenSuse:
sudo chown -R root:adm /var/log/swift sudo chmod -R g+w /var/log/swift sudo systemctl restart rsyslog sudo systemctl enable rsyslog
Configuring each node
After performing the following steps, be sure to verify that Swift has access to resulting configuration files (sample configuration files are provided with all defaults in line-by-line comments).
Optionally remove an existing swift directory:
sudo rm -rf /etc/swiftPopulate the
/etc/swiftdirectory itself:cd $HOME/swift/doc; sudo cp -r saio/swift /etc/swift; cd - sudo chown -R ${USER}:${USER} /etc/swiftUpdate
<your-user-name>references in the Swift config files:find /etc/swift/ -name \*.conf | xargs sudo sed -i "s/<your-user-name>/${USER}/"
The contents of the configuration files provided by executing the above commands are as follows:
/etc/swift/swift.conf/../saio/swift/swift.conf
/etc/swift/proxy-server.conf/../saio/swift/proxy-server.conf
/etc/swift/object-expirer.conf/../saio/swift/object-expirer.conf
/etc/swift/container-reconciler.conf/../saio/swift/container-reconciler.conf
/etc/swift/container-sync-realms.conf/../saio/swift/container-sync-realms.conf
/etc/swift/account-server/1.conf/../saio/swift/account-server/1.conf
/etc/swift/container-server/1.conf/../saio/swift/container-server/1.conf
/etc/swift/object-server/1.conf/../saio/swift/object-server/1.conf
/etc/swift/account-server/2.conf/../saio/swift/account-server/2.conf
/etc/swift/container-server/2.conf/../saio/swift/container-server/2.conf
/etc/swift/object-server/2.conf/../saio/swift/object-server/2.conf
/etc/swift/account-server/3.conf/../saio/swift/account-server/3.conf
/etc/swift/container-server/3.conf/../saio/swift/container-server/3.conf
/etc/swift/object-server/3.conf/../saio/swift/object-server/3.conf
/etc/swift/account-server/4.conf/../saio/swift/account-server/4.conf
/etc/swift/container-server/4.conf/../saio/swift/container-server/4.conf
/etc/swift/object-server/4.conf/../saio/swift/object-server/4.conf
Setting up scripts for running Swift
Copy the SAIO scripts for resetting the environment:
mkdir -p $HOME/bin cd $HOME/swift/doc; cp saio/bin/* $HOME/bin; cd - chmod +x $HOME/bin/*Edit the
$HOME/bin/resetswiftscriptThe template
resetswiftscript looks like the following:/../saio/bin/resetswift
If you did not set up rsyslog for individual logging, remove the
find /var/log/swift...line:sed -i "/find \/var\/log\/swift/d" $HOME/bin/resetswiftInstall the sample configuration file for running tests:
cp $HOME/swift/test/sample.conf /etc/swift/test.confThe template
test.conflooks like the following:/../../test/sample.conf
Configure environment variables for Swift
Add an environment variable for running tests below:
echo "export SWIFT_TEST_CONFIG_FILE=/etc/swift/test.conf" >> $HOME/.bashrcBe sure that your
PATHincludes thebindirectory:echo "export PATH=${PATH}:$HOME/bin" >> $HOME/.bashrcIf you are using a loopback device for Swift Storage, add an environment var to substitute
/dev/sdb1with/srv/swift-disk:echo "export SAIO_BLOCK_DEVICE=/srv/swift-disk" >> $HOME/.bashrcIf you are using a device other than
/dev/sdb1for Swift storage (for example,/dev/vdb1), add an environment var to substitute it:echo "export SAIO_BLOCK_DEVICE=/dev/vdb1" >> $HOME/.bashrcIf you are using a location other than
/tmpfor Swift tmp data (for example,/mnt/tmp), addTMPDIRenvironment var to set it:export TMPDIR=/mnt/tmp echo "export TMPDIR=/mnt/tmp" >> $HOME/.bashrcSource the above environment variables into your current environment:
. $HOME/.bashrc
Constructing initial rings
Construct the initial rings using the provided script:
remakeringsThe
remakeringsscript looks like the following:/../saio/bin/remakerings
You can expect the output from this command to produce the following. Note that 3 object rings are created in order to test storage policies and EC in the SAIO environment. The EC ring is the only one with all 8 devices. There are also two replication rings, one for 3x replication and another for 2x replication, but those rings only use 4 devices:
Device d0r1z1-127.0.0.1:6010R127.0.0.1:6010/sdb1_"" with 1.0 weight got id 0 Device d1r1z2-127.0.0.2:6020R127.0.0.2:6020/sdb2_"" with 1.0 weight got id 1 Device d2r1z3-127.0.0.3:6030R127.0.0.3:6030/sdb3_"" with 1.0 weight got id 2 Device d3r1z4-127.0.0.4:6040R127.0.0.4:6040/sdb4_"" with 1.0 weight got id 3 Reassigned 3072 (300.00%) partitions. Balance is now 0.00. Dispersion is now 0.00 Device d0r1z1-127.0.0.1:6010R127.0.0.1:6010/sdb1_"" with 1.0 weight got id 0 Device d1r1z2-127.0.0.2:6020R127.0.0.2:6020/sdb2_"" with 1.0 weight got id 1 Device d2r1z3-127.0.0.3:6030R127.0.0.3:6030/sdb3_"" with 1.0 weight got id 2 Device d3r1z4-127.0.0.4:6040R127.0.0.4:6040/sdb4_"" with 1.0 weight got id 3 Reassigned 2048 (200.00%) partitions. Balance is now 0.00. Dispersion is now 0.00 Device d0r1z1-127.0.0.1:6010R127.0.0.1:6010/sdb1_"" with 1.0 weight got id 0 Device d1r1z1-127.0.0.1:6010R127.0.0.1:6010/sdb5_"" with 1.0 weight got id 1 Device d2r1z2-127.0.0.2:6020R127.0.0.2:6020/sdb2_"" with 1.0 weight got id 2 Device d3r1z2-127.0.0.2:6020R127.0.0.2:6020/sdb6_"" with 1.0 weight got id 3 Device d4r1z3-127.0.0.3:6030R127.0.0.3:6030/sdb3_"" with 1.0 weight got id 4 Device d5r1z3-127.0.0.3:6030R127.0.0.3:6030/sdb7_"" with 1.0 weight got id 5 Device d6r1z4-127.0.0.4:6040R127.0.0.4:6040/sdb4_"" with 1.0 weight got id 6 Device d7r1z4-127.0.0.4:6040R127.0.0.4:6040/sdb8_"" with 1.0 weight got id 7 Reassigned 6144 (600.00%) partitions. Balance is now 0.00. Dispersion is now 0.00 Device d0r1z1-127.0.0.1:6011R127.0.0.1:6011/sdb1_"" with 1.0 weight got id 0 Device d1r1z2-127.0.0.2:6021R127.0.0.2:6021/sdb2_"" with 1.0 weight got id 1 Device d2r1z3-127.0.0.3:6031R127.0.0.3:6031/sdb3_"" with 1.0 weight got id 2 Device d3r1z4-127.0.0.4:6041R127.0.0.4:6041/sdb4_"" with 1.0 weight got id 3 Reassigned 3072 (300.00%) partitions. Balance is now 0.00. Dispersion is now 0.00 Device d0r1z1-127.0.0.1:6012R127.0.0.1:6012/sdb1_"" with 1.0 weight got id 0 Device d1r1z2-127.0.0.2:6022R127.0.0.2:6022/sdb2_"" with 1.0 weight got id 1 Device d2r1z3-127.0.0.3:6032R127.0.0.3:6032/sdb3_"" with 1.0 weight got id 2 Device d3r1z4-127.0.0.4:6042R127.0.0.4:6042/sdb4_"" with 1.0 weight got id 3 Reassigned 3072 (300.00%) partitions. Balance is now 0.00. Dispersion is now 0.00Read more about Storage Policies and your SAIO
policies_saio
Testing Swift
Verify the unit tests run:
$HOME/swift/.unittestsNote that the unit tests do not require any Swift daemons running.
Start the "main" Swift daemon processes (proxy, account, container, and object):
startmain(The "
Unable to increase file descriptor limit. Running as non-root?" warnings are expected and ok.)The
startmainscript looks like the following:/../saio/bin/startmain
Get an
X-Storage-UrlandX-Auth-Token:curl -v -H 'X-Storage-User: test:tester' -H 'X-Storage-Pass: testing' http://127.0.0.1:8080/auth/v1.0Check that you can
GETaccount:curl -v -H 'X-Auth-Token: <token-from-x-auth-token-above>' <url-from-x-storage-url-above>Check that
swiftcommand provided by the python-swiftclient package works:swift -A http://127.0.0.1:8080/auth/v1.0 -U test:tester -K testing statVerify the functional tests run:
$HOME/swift/.functests(Note: functional tests will first delete everything in the configured accounts.)
Verify the probe tests run:
$HOME/swift/.probetests(Note: probe tests will reset your environment as they call
resetswiftfor each test.)
Debugging Issues
If all doesn't go as planned, and tests fail, or you can't auth, or something doesn't work, here are some good starting places to look for issues:
- Everything is logged using system facilities -- usually in
/var/log/syslog, but possibly in/var/log/messageson e.g. Fedora -- so that is a good first place to look for errors (most likely python tracebacks). - Make sure all of the server processes are running. For the base functionality, the Proxy, Account, Container, and Object servers should be running.
- If one of the servers are not running, and no errors are logged to
syslog, it may be useful to try to start the server manually, for
example:
swift-object-server /etc/swift/object-server/1.confwill start the object server. If there are problems not showing up in syslog, then you will likely see the traceback on startup. - If you need to, you can turn off syslog for unit tests. This can be
useful for environments where
/dev/logis unavailable, or which cannot rate limit (unit tests generate a lot of logs very quickly). Open the fileSWIFT_TEST_CONFIG_FILEpoints to, and change the value offake_syslogtoTrue. - If you encounter a
401 Unauthorizedwhen following Step 12 where you check that you canGETaccount, usesudo service memcached statusand check if memcache is running. If memcache is not running, start it usingsudo service memcached start. Once memcache is running, rerunGETaccount.
Known Issues
Listed here are some "gotcha's" that you may run into when using or testing your SAIO:
- fallocate_reserve - in most cases a SAIO doesn't have a very large XFS partition so having fallocate enabled and fallocate_reserve set can cause issues, specifically when trying to run the functional tests. For this reason fallocate has been turned off on the object-servers in the SAIO. If you want to play with the fallocate_reserve settings then know that functional tests will fail unless you change the max_file_size constraint to something more reasonable then the default (5G). Ideally you'd make it 1/4 of your XFS file system size so the tests can pass.