DAOS Set-Up on OpenSUSE¶

Introduction¶

The following instructions detail how to install, set up and start DAOS servers and clients on two or more nodes. This document includes instructions for openSUSE/SLES-compatible distributions.

For setup instructions on RHEL and RHEL clones, refer to the RHEL setup.

For more details, including the prerequisite steps before installing DAOS, reference the DAOS administration guide.

Requirements¶

The following steps require two or more hosts which will be divided up into admin, client, and server roles. One node can be used for multiple roles. For example the admin role can reside on a server, on a client, or on a dedicated admin node.

All nodes must have:

sudo access configured
password-less ssh configured
a parallel command launcher like pdsh or clush is installed

In addition the server nodes should also have:

For the use of the commands outlined on this page the following shell variables will need to be defined:

ADMIN_NODES
CLIENT_NODES
SERVER_NODES
ALL_NODES

For example, if you want to use admin-1 as the admin node, client-1 and client-2 as client nodes, and server-[1-3] as server nodes, these variables would be defined as:

ADMIN_NODES="admin-1"

CLIENT_NODES="client-1,client-2"

SERVER_NODES="server-1,server-2"

ALL_NODES="$ADMIN_NODES,$CLIENT_NODES,$SERVER_NODES"

Note

If a client node is also serving as an admin node, exclude $ADMIN_NODES from the ALL_NODES assignment to prevent duplication. For example: ALL_NODES=$CLIENT_NODES,$SERVER_NODES

RPM Installation¶

In this section the required RPMs will be installed on each of the nodes based upon their role. Admin and client nodes require the installation of the daos-client RPM and the server nodes require the installation of the daos-server RPM.

Configure access to the DAOS package repository:

clush -B -w $ALL_NODES 'sudo zypper ar https://packages.daos.io/v2.6/Leap15/packages/x86_64/daos_packages.repo'

Import GPG key on all nodes:

clush -B -w $ALL_NODES 'sudo rpm --import https://packages.daos.io/RPM-GPG-KEY'

Refresh zypper:

clush -B -w $ALL_NODES 'sudo zypper --non-interactive refresh'

Install mercury-ucx or mercury-libfabric RPMs on all nodes. For more details, reference Network Requirements.
```
clush -B -w $ALL_NODES 'sudo zypper install -y mercury-ucx'
```

Install the daos-admin RPMs on the admin nodes:

clush -B -w $ADMIN_NODES 'sudo zypper install -y daos-admin'

Install the daos-server RPMs on the server nodes:

clush -B -w $SERVER_NODES 'sudo zypper install -y daos-server'

Install the daos-client RPMs on the client nodes:

clush -B -w $CLIENT_NODES 'sudo zypper install -y daos-client'

Generate certificates¶

In this section certificates will be generated and installed for encrypting the DAOS control plane communications.

Administrative nodes require the following certificate files:

CA root certificate (daosCA.crt) owned by the current user
Admin certificate (admin.crt) owned by the current user
Admin key (admin.key) owned by the current user

Client nodes require the following certificate files:

CA root certificate (daosCa.crt) owned by the daos_agent user
Agent certificate (agent.crt) owned by the daos_agent user
Agent key (agent.key) owned by the daos_agent user

Server nodes require the following certificate files:

CA root certificate (daosCA.crt) owned by the daos_server user
Server certificate (server.crt) owned by the daos_server user
Server key (server.key) owned by the daos_server user
A copy of the Client certificate (client.crt) owned by the daos_server user

See Certificate Configuration for more information.

Note

The following commands are run on one of the $ADMIN_NODES.

Generate a new set of certificates:
```
cd /tmp
/usr/lib64/daos/certgen/gen_certificates.sh
```
Note

These files should be protected from unauthorized access and preserved for future use.
Copy the certificates to a common location on each node in order to move them to the final location:
```
clush -B -w $ALL_NODES --copy /tmp/daosCA --dest /tmp
```

Copy the certificates to their default location (/etc/daos) on each admin node:

clush -B -w $ADMIN_NODES sudo cp /tmp/daosCA/certs/daosCA.crt /etc/daos/certs/.
clush -B -w $ADMIN_NODES sudo cp /tmp/daosCA/certs/admin.crt /etc/daos/certs/.
clush -B -w $ADMIN_NODES sudo cp /tmp/daosCA/certs/admin.key /etc/daos/certs/.

Note

If the /etc/daos/certs directory does not exist on the admin nodes then use the following command to create it:

    clush -B -w $ADMIN_NODES sudo mkdir /etc/daos/certs

Copy the certificates to their default location (/etc/daos) on each client node:

clush -B -w $CLIENT_NODES sudo cp /tmp/daosCA/certs/daosCA.crt /etc/daos/certs/.
clush -B -w $CLIENT_NODES sudo cp /tmp/daosCA/certs/agent.crt /etc/daos/certs/.
clush -B -w $CLIENT_NODES sudo cp /tmp/daosCA/certs/agent.key /etc/daos/certs/.

Note

If the /etc/daos/certs directory does not exist on the client nodes, use the following command to create it:

clush -B -w $CLIENT_NODES sudo mkdir /etc/daos/certs

Copy the certificates to their default location (/etc/daos) on each server node:

clush -B -w $SERVER_NODES sudo cp /tmp/daosCA/certs/daosCA.crt /etc/daos/certs/.
clush -B -w $SERVER_NODES sudo cp /tmp/daosCA/certs/server.crt /etc/daos/certs/.
clush -B -w $SERVER_NODES sudo cp /tmp/daosCA/certs/server.key /etc/daos/certs/.
clush -B -w $SERVER_NODES sudo cp /tmp/daosCA/certs/agent.crt /etc/daos/certs/clients/agent.crt

Cleanup the temp directory

clush -B -w $ALL_NODES sudo rm -rf /tmp/daosCA

Set the ownership of the admin certificates on each admin node:

clush -B -w $ADMIN_NODES sudo chown $USER: /etc/daos/certs/daosCA.crt
clush -B -w $ADMIN_NODES sudo chown $USER: /etc/daos/certs/admin.*

Set the ownership of the client certificates on each client node:

clush -B -w $CLIENT_NODES sudo chown $USER: /etc/daos/certs/daosCA.crt
clush -B -w $CLIENT_NODES sudo chown daos_agent:daos_agent /etc/daos/certs/agent.*

Set the ownership of the server certificates on each server node:

clush -B -w $SERVER_NODES sudo chown daos_server:daos_server /etc/daos/certs/daosCA.crt
clush -B -w $SERVER_NODES sudo chown daos_server:daos_server /etc/daos/certs/server.*
clush -B -w $SERVER_NODES sudo chown daos_server:daos_server /etc/daos/certs/clients/agent.crt
clush -B -w $SERVER_NODES sudo chown daos_server:daos_server /etc/daos/certs/clients

Hardware Provisioning¶

If the server nodes are configured with PMem (Intel(R) Optane(TM) persistent memory), it will need to be prepared and configured to be used by DAOS and NVME SSDs will be identified. See SCM Preparation for more infortamtion.

Prepare the pmem devices on Server nodes:

daos_server scm prepare

Sample Script:

Prepare locally-attached PMem\...

Memory allocation goals for PMem will be changed and namespaces
modified, this may be a destructive operation. Please ensure
namespaces are unmounted and locally attached PMem modules are
not in use. Please be patient as it may take several minutes and
subsequent reboot maybe required.

Are you sure you want to continue? (yes/no)

yes

A reboot is required to process new PMem memory allocation goals.

Reboot the server node.

Run the prepare cmdline again:

daos_server scm prepare

Sample Script:

Prepare locally-attached PMem\...
SCM namespaces:
SCM Namespace   Socket ID   Capacity
-------------   ---------   --------
pmem0           0           3.2 TB
pmem1           0           3.2 TB

Scan the available scm storage on the Server nodes:

daos_server scm scan
SCM Namespace   Socket ID   Capacity
-------------   ---------   --------
pmem0           0           3.2 TB
pmem1           1           3.2 TB

Create Configuration Files¶

In this section the daos_server, daos_agent, and dmg command configuration files will be defined. Examples are available on github.

Determine the addresses for the NVMe storage on the server nodes:

clush -B -w $SERVER_NODES daos_server nvme scan --ignore-config

Sample daos_server nvme scan output:

Scan locally-attached NVMe storage...
NVMe PCI     Model               FW Revision Socket Capacity Role(s) Rank
--------     -----               ----------- ------ -------- ------- ----
0000:83:00.0 INTEL SSDPE2MD800G4 8DV10171    0      800 GB   NA      None
0000:84:00.0 INTEL SSDPE2MD800G4 8DV10171    1      800 GB   NA      None

Note

Save the addresses of the NVMe devices to use with each DAOS server, e.g. "81:00.0", from each server node. This information will be used to populate the "bdev_list" server configuration parameter below.

Create a server configuration file. See DAOS Server Setup for more details. Either modify the sample /etc/daos/daos_server.yml file or use the config generate tool to create a server configuration file in /tmp/daos_server.yml:

An example of using the config generate command on the first server node:

daos_server config generate --access-points=$(hostname -s) | tee /tmp/daos_server.yml

An example of modifying the sample daos_server.yml:

cp /etc/daos/daos_server.yml /tmp/daos_server.yml
vim /tmp/daos_server.yml

Note

Use any PMEM or NVMe addresses collected above when modifying the scm_list or bdev_list engine storage entries, respectively.

An example of the /tmp/daos_server.yml:

cat /tmp/daos_server.yml
name: daos_server
access_points:
- node-4
port: 10001

transport_config:
    allow_insecure: false
    client_cert_dir: /etc/daos/certs/clients
    ca_cert: /etc/daos/certs/daosCA.crt
    cert: /etc/daos/certs/server.crt
    key: /etc/daos/certs/server.key
provider: ofi+verbs;ofi_rxm
control_log_mask: DEBUG
control_log_file: /tmp/daos_server.log
helper_log_file: /tmp/daos_server_helper.log
engines:
-
    pinned_numa_node: 0
    targets: 8
    nr_xs_helpers: 2
    fabric_iface: ib0
    fabric_iface_port: 31316
    log_mask: INFO
    log_file: /tmp/daos_engine_0.log
    env_vars:
        - CRT_TIMEOUT=30
    storage:
    -
        class: dcpm
        scm_mount: /mnt/daos0
        scm_list:
        - /dev/pmem0
    -
        class: nvme
        bdev_list:
        - "0000:81:00.0"
-
    pinned_numa_node: 1
    targets: 8
    nr_xs_helpers: 2
    fabric_iface: ib1
    fabric_iface_port: 31416
    log_mask: INFO
    log_file: /tmp/daos_engine_1.log
    env_vars:
        - CRT_TIMEOUT=30
    storage:
    -
        class: dcpm
        scm_mount: /mnt/daos1
        scm_list:
        - /dev/pmem1
    -
        class: nvme
        bdev_list:
        - "0000:83:00.0"

Copy the modified server yaml file to all the server nodes at /etc/daos/daos_server.yml.

clush -B -w $SERVER_NODES --copy /tmp/daos_server.yml
clush -B -w $SERVER_NODES sudo cp /tmp/daos_server.yml /etc/daos/

Create an agent configuration file by modifying the default /etc/daos/daos_agent.yml file on the client nodes. The following is an example daos_agent.yml. Copy the modified agent yaml file to all the client nodes at /etc/daos/daos_agent.yml.

cat /tmp/daos_agent.yml
name: daos_server
access_points:
- server-1

port: 10001

transport_config:
    allow_insecure: false
    ca_cert: /etc/daos/certs/daosCA.crt
    cert: /etc/daos/certs/agent.crt
    key: /etc/daos/certs/agent.key
log_file: /tmp/daos_agent.log

clush -B -w $CLIENT_NODES --copy /tmp/daos_agent.yml
clush -B -w $CLIENT_NODES sudo cp /tmp/daos_agent.yml /etc/daos/

Create a dmg configuration file by modifying the default /etc/daos/daos_control.yml file on the admin node. The following is an example of the daos_control.yml.

cat /tmp/daos_control.yml
name: daos_server
port: 10001
hostlist:
- server-1
- server-2

transport_config:
    allow_insecure: false
    ca_cert: /etc/daos/certs/daosCA.crt
    cert: /etc/daos/certs/admin.crt
    key: /etc/daos/certs/admin.key

clush -B -w $ADMIN_NODES --copy /tmp/daos_control.yml
clush -B -w $ADMIN_NODES sudo cp /tmp/daos_control.yml /etc/daos/

Start the DAOS Servers¶

Start daos engines on server nodes:

clush -B -w $SERVER_NODES "sudo systemctl daemon-reload"
clush -B -w $SERVER_NODES "sudo systemctl start daos_server"

Check status and format storage:

# check status
clush -B -w $SERVER_NODES "sudo systemctl status daos_server"

# if you see following format messages (depending on number of servers), proceed to storage format
server-1: server-1.test.example.com INFO 2023/04/11 23:14:06 SCM format required on instance 1
server-1: server-1.test.example.com INFO 2023/04/11 23:14:06 SCM format required on instance 0

# format storage
dmg storage format -l $SERVER_NODES # can use --force if needed

Verify that all servers have started:

# system query from ADMIN_NODES
dmg system query -v

# all the server ranks should show 'Joined' STATE
Rank UUID                                 Control Address  Fault Domain                  State  Reason
---- ----                                 ---------------  ------------                  -----  ------
0    604c4ffa-563a-49dc-b702-3c87293dbcf3 10.8.1.179:10001 /server-1.test.example.com Joined
1    f0791f98-4379-4ace-a083-6ca3ffa65756 10.8.1.179:10001 /server-1.test.example.com Joined
2    745d2a5b-46dd-42c5-b90a-d2e46e178b3e 10.8.1.189:10001 /server-2.test.example.com Joined
3    ba6a7800-3952-46ce-af92-bba9daa35048 10.8.1.189:10001 /server-2.test.example.com Joined

Start the DAOS Agents¶

Start the daos agents on the client nodes:

# start agents
clush -B -w $CLIENT_NODES "sudo systemctl start daos_agent"

Verify daos_agent communication:

# verify client communication to each server rank
clush -B -w $CLIENT_NODES "daos system query --verbose"

# Sample output
connected to DAOS system:
name: daos_server
fabric provider: ofi+tcp
access point ranks:
        rank[0]: ofi+tcp://10.8.1.179:10001
rank URIs:
        rank[0]: ofi+tcp://10.8.1.179:10001
        rank[1]: ofi+tcp://10.8.1.179:10001
        rank[2]: ofi+tcp://10.8.1.189:10001
        rank[3]: ofi+tcp://10.8.1.189:10001