You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 4b80f24
Browse filesBrowse the repository at this point in the historyBrowse files
Obstor is a high-performance object storage system supporting popular transfer protocols like S3 and SFTP, making it suitable for building high-performance infrastructure for machine learning, analytics, and application data workloads. Obstor is based on the 2021 Apache-licensed release of MinIO, prior to the project's transition to AGPL and later archival.
4
+
Obstor is a high-performance object storage system supporting popular transfer protocols like S3 and SFTP, making it suitable for building high-performance infrastructure for machine learning, analytics, and application data workloads. Obstor is based on the 2021 Apache-licensed release of Obstor, prior to the project's transition to AGPL and later archival.
7
5
8
6
This README provides quickstart instructions on running Obstor on baremetal hardware, including Docker-based installations. For Kubernetes environments,
9
-
use the [Obstor Kubernetes Operator](https://github.com/minio/operator/blob/master/README.md).
7
+
use the [Obstor Kubernetes Operator](https://github.com/obstor/operator/blob/master/README.md).
10
8
11
9
# Docker Installation
12
10
@@ -22,15 +20,15 @@ for more complete documentation.
22
20
Run the following command to run the latest stable image of Obstor on a Docker container using an ephemeral data volume:
23
21
24
22
```sh
25
-
docker run -p 9000:9000 ghcr.io/cloudment/obstor server /data
23
+
docker run -p 9000:9000 ghcr.io/obstor/obstor server /data
26
24
```
27
25
28
26
The Obstor deployment starts using default root credentials `obstoradmin:obstoradmin`. You can test the deployment using the Obstor Browser, an embedded
29
27
web-based object browser built into Obstor Server. Point a web browser running on the host machine to http://127.0.0.1:9000 and log in with the
30
28
root credentials. You can use the Browser to create buckets, upload objects, and browse the contents of the Obstor server.
31
29
32
-
You can also connect using any S3-compatible tool, such as the Obstor Client `mc` commandline tool. See
33
-
[Test using Obstor Client `mc`](#test-using-obstor-client-mc) for more information on using the `mc` commandline tool. For application developers,
30
+
You can also connect using any S3-compatible tool, such as `rclone`, `rsync` (over an S3 mount), or the AWS CLI. See
31
+
[Test using an S3 client](#test-using-an-s3-client) for more information. For application developers,
34
32
see https://obstor.net/docs/ and click **OBSTOR SDKS** in the navigation to view Obstor SDKs for supported languages.
35
33
36
34
@@ -71,8 +69,8 @@ The Obstor deployment starts using default root credentials `obstoradmin:obstora
71
69
web-based object browser built into Obstor Server. Point a web browser running on the host machine to http://127.0.0.1:9000 and log in with the
72
70
root credentials. You can use the Browser to create buckets, upload objects, and browse the contents of the Obstor server.
73
71
74
-
You can also connect using any S3-compatible tool, such as the Obstor Client `mc` commandline tool. See
75
-
[Test using Obstor Client `mc`](#test-using-obstor-client-mc) for more information on using the `mc` commandline tool. For application developers,
72
+
You can also connect using any S3-compatible tool, such as `rclone`, `rsync` (over an S3 mount), or the AWS CLI. See
73
+
[Test using an S3 client](#test-using-an-s3-client) for more information. For application developers,
76
74
see https://obstor.net/docs/ and click **OBSTOR SDKS** in the navigation to view Obstor SDKs for supported languages.
77
75
78
76
@@ -99,8 +97,8 @@ The Obstor deployment starts using default root credentials `obstoradmin:obstora
99
97
web-based object browser built into Obstor Server. Point a web browser running on the host machine to http://127.0.0.1:9000 and log in with the
100
98
root credentials. You can use the Browser to create buckets, upload objects, and browse the contents of the Obstor server.
101
99
102
-
You can also connect using any S3-compatible tool, such as the Obstor Client `mc` commandline tool. See
103
-
[Test using Obstor Client `mc`](#test-using-obstor-client-mc) for more information on using the `mc` commandline tool. For application developers,
100
+
You can also connect using any S3-compatible tool, such as `rclone`, `rsync` (over an S3 mount), or the AWS CLI. See
101
+
[Test using an S3 client](#test-using-an-s3-client) for more information. For application developers,
104
102
see https://obstor.net/docs/ and click **OBSTOR SDKS** in the navigation to view Obstor SDKs for supported languages.
105
103
106
104
> NOTE: Standalone Obstor servers are best suited for early development and evaluation. Certain features such as versioning, object locking, and bucket replication
@@ -120,8 +118,8 @@ The Obstor deployment starts using default root credentials `obstoradmin:obstora
120
118
web-based object browser built into Obstor Server. Point a web browser running on the host machine to http://127.0.0.1:9000 and log in with the
121
119
root credentials. You can use the Browser to create buckets, upload objects, and browse the contents of the Obstor server.
122
120
123
-
You can also connect using any S3-compatible tool, such as the Obstor Client `mc` commandline tool. See
124
-
[Test using Obstor Client `mc`](#test-using-obstor-client-mc) for more information on using the `mc` commandline tool. For application developers,
121
+
You can also connect using any S3-compatible tool, such as `rclone`, `rsync` (over an S3 mount), or the AWS CLI. See
122
+
[Test using an S3 client](#test-using-an-s3-client) for more information. For application developers,
125
123
see https://obstor.net/docs/ and click **OBSTOR SDKS** in the navigation to view Obstor SDKs for supported languages.
126
124
127
125
@@ -199,38 +197,34 @@ The above statement is also valid for all gateway backends.
199
197
## Test using Browser Dashboard
200
198
Obstor Server comes with an embedded web based object browser. Point your web browser to http://127.0.0.1:9000 to ensure your server has started successfully.
`mc` provides a modern alternative to UNIX commands like ls, cat, cp, mirror, diff etc. It supports filesystems and Amazon S3-compatible cloud storage services. Follow the Obstor Client [Quickstart Guide](https://obstor.net/docs/obstor-client-quickstart-guide) for further instructions.
204
+
## Test using an S3 client
205
+
Obstor is S3-compatible, so any S3 client works for data access. `rclone` is a modern alternative to UNIX commands like ls, cat, cp, sync, etc. against S3-compatible storage; `rsync` works over an S3 mount (for example an `rclone mount` or `s3fs` mount); the AWS CLI (`aws s3`) also works. Point the client at the server endpoint with a root or user credential.
208
206
209
207
# Upgrading Obstor
210
-
Obstor server supports rolling upgrades, i.e. you can update one Obstor instance at a time in a distributed cluster. This allows upgrades with no downtime. Upgrades can be done manually by replacing the binary with the latest release and restarting all servers in a rolling fashion. However, we recommend all our users to use [`mc admin update`](https://obstor.net/docs/obstor-admin-complete-guide#update) from the client. This will update all the nodes in the cluster simultaneously and restart them, as shown in the following command from the Obstor client (mc):
211
-
212
-
```
213
-
mc admin update <obstor alias, e.g., myobstor>
214
-
```
208
+
Obstor server supports rolling upgrades, i.e. you can update one Obstor instance at a time in a distributed cluster. This allows upgrades with no downtime. Upgrade by replacing the binary with the latest release and restarting servers one at a time in a rolling fashion.
215
209
216
-
> NOTE: some releases might not allow rolling upgrades, this is always called out in the release notes and it is generally advised to read release notes before upgrading. In such a situation `mc admin update` is the recommended upgrading mechanism to upgrade all servers at once.
210
+
> NOTE: some releases might not allow rolling upgrades, this is always called out in the release notes and it is generally advised to read release notes before upgrading.
217
211
218
212
## Important things to remember during Obstor upgrades
219
213
220
-
-`mc admin update` will only work if the user running Obstor has write access to the parent directory where the binary is located, for example if the current binary is at `/usr/local/bin/obstor`, you would need write access to `/usr/local/bin`.
221
-
-`mc admin update` updates and restarts all servers simultaneously, applications would retry and continue their respective operations upon upgrade.
222
-
-`mc admin update` is disabled in kubernetes/container environments, container environments provide their own mechanisms to rollout of updates.
223
-
- In the case of federated setups`mc admin update` should be run against each cluster individually. Avoid updating `mc` to any new releases until all clusters have been successfully updated.
224
-
- If using `kes` as KMS with Obstor, just replace the binary and restart `kes` more information about `kes` can be found [here](https://github.com/minio/kes/wiki)
214
+
-Replacing the binary only works if the user running Obstor has write access to the parent directory where the binary is located, for example if the current binary is at `/usr/local/bin/obstor`, you would need write access to `/usr/local/bin`.
215
+
-Restart servers one at a time; applications retry and continue their respective operations during the rolling restart.
216
+
-In kubernetes/container environments, use the platform's own mechanism to roll out the updated image.
217
+
- In the case of federated setups, upgrade each cluster individually.
218
+
- If using `kes` as KMS with Obstor, just replace the binary and restart `kes` more information about `kes` can be found [here](https://github.com/obstor/kes/wiki)
225
219
- If using Vault as KMS with Obstor, ensure you have followed the Vault upgrade procedure outlined here: https://www.vaultproject.io/docs/upgrading/index.html
226
220
- If using etcd with Obstor for the federation, ensure you have followed the etcd upgrade procedure outlined here: https://github.com/etcd-io/etcd/blob/master/Documentation/upgrades/upgrading-etcd.md
0 commit comments