Remove version details and history from doc files #22
Conversation
|
@hwware Done, updated the diff. Please have a look. |
Signed-off-by: Shivshankar-Reddy <[email protected]>
Signed-off-by: Shivshankar-Reddy <[email protected]>
Signed-off-by: Shivshankar-Reddy <[email protected]>
Signed-off-by: Shivshankar-Reddy <[email protected]>
Shivshankar-Reddy
force-pushed
the
Rename-version
branch
from
15edd5a
to
33f1be5
Compare
| As of Redis version 2.6.0, scripts were replicated verbatim, meaning that the scripts' source code was sent for execution by replicas and stored in the AOF. | ||
| An alternative replication mode added in version 3.2.0 allows replicating only the scripts' effects. | ||
| As of Redis version 7.0, script replication is no longer supported, and the only replication mode available is script effects replication. | ||
| Verbatim script replication, meaning that the scripts' source code was sent for execution by replicas and stored in the AOF. |
There was a problem hiding this comment.
I think we should rephrase this part statement, Since Redis 7.0, verbatim replication is no longer supported, I think we do not need to mention verbatim replication anymore
Signed-off-by: Shivshankar-Reddy <[email protected]>
| as it does in other Linux based environments. This is because we believe | ||
| that for the small use cases inside embedded devices, memory fragmentation | ||
| is unlikely to be a problem. Moreover `jemalloc` on ARM may not be as tested | ||
| as the `libc` allocator. | ||
|
|
||
| ## Performance |
There was a problem hiding this comment.
@stockholmux Removed this section as performance test done way long back on redis 4.0, I am not sure whether they are really relevant now. Hopefully, someone will do the test and based on Valkey actual performance results on PI, we can bring this section back.
|
@hwware @stockholmux I have updated the diff based on your suggestions, Can you please have a loo when have some time? |
Signed-off-by: Shivshankar-Reddy <[email protected]>
|
Now it is good to me. I approve it. |
| * `pubsub_clients`: Number of clients in pubsub mode (`SUBSCRIBE`, `PSUBSCRIBE`, `SSUBSCRIBE`). Added in Valkey 8.0 | ||
| * `watching_clients`: Number of clients in watching mode (`WATCH`). Added in Valkey 8.0 | ||
| * `pubsub_clients`: Number of clients in pubsub mode (`SUBSCRIBE`, `PSUBSCRIBE`, `SSUBSCRIBE`). | ||
| * `watching_clients`: Number of clients in watching mode (`WATCH`). | ||
| * `clients_in_timeout_table`: Number of clients in the clients timeout table | ||
| * `total_watched_keys`: Number of watched keys. Added in Valkey 8.0. | ||
| * `total_watched_keys`: Number of watched keys. |
There was a problem hiding this comment.
I'm strongly against this change.
Where is the decision to remove all history? Please show me.
It is a terrible idea for anyone who is upgrading from Redis, which I think will be 99% of our users.
I insist that we keep references to Redis 6.2 and later, i.e. the versions that were still supported when we forked, which is what all users are running before they upgrade to Valkey.
@valkey-io/core-team I would like to know where this idea is coming from. Was it discussed anywhere?
There was a problem hiding this comment.
yes, i think we should keep the history, finding the corresponding supported version is a pain (it deserves to be properly documented at somewhere). I don't remember us discussing this.
There was a problem hiding this comment.
We didn't discuss this in the core team meeting but there was a discussion on an early re-branding PR (and this predates @enjoy-binbin's time on this project).
IIRC, the decision back then is to keep anything beyond 7.2 (the origin of Valkey) and this is why 8.0 was kept.
I insist that we keep references to Redis 6.2 and later
I am not against keeping references to Redis 6.2+ but can you explain how it could help customers who switch over to Valkey?
There was a problem hiding this comment.
Some users avoid upgrading until it is absolutely necessary. When a version is end of life, users want to switch to the latest version. That means from redis 6.2 to valkey 7.2 or valkey 8.
There was a problem hiding this comment.
I can't see in that PR that there is a decision to keep anything beyond 7.2.
- Some really old stuff is deleted, like redis 2 (that's good)
- Valkey 8 is kept. This is good and important, since it is a future version and not yet released. It's important to keep this to indicate that these parts are not valid for our first release 7.2.5.
- Delete references to 7.2, 7.0, 6.2 is unfortunate but no disaster. I think we should avoid it. We don't have versioned docs so these indicators is the only way for users to find these differences when upgrading, apart from the release notes in each release.
There was a problem hiding this comment.
Some users avoid upgrading until it is absolutely necessary. When a version is end of life, users want to switch to the latest version. That means from redis 6.2 to valkey 7.2 or valkey 8.
Got it - I think we should re-instate the history of these versions then. I am all for helping the users.
| @@ -681,7 +658,6 @@ In addition, the following external libraries are loaded and accessible to scrip | |||
|
|
|||
| ### <a name="os-library"></a> _os_ library | |||
|
|
|||
| * Since version: 8.0.0 | |||
| @@ -361,8 +347,6 @@ If you run this script by calling `EVAL "..." 3 A B C 1 2 3`, the result will be | |||
|
|
|||
| ### <a name="redis.replicate_commands"></a> `redis.replicate_commands()` | |||
There was a problem hiding this comment.
Ideally just delete this, it's no longer supported.
Documents the behavior for the new server top level domain object and the new Lua fields. Also updates all of the examples to use the server object. Also includes some cleanup to references to "Valkey serialization protocol" which doesn't exist. There will be some minimal conflicts with #22, as it attempts to remove the versioning references which this file also does for the files that were touched. --------- Signed-off-by: Madelyn Olson <[email protected]> Signed-off-by: Viktor Söderqvist <[email protected]> Co-authored-by: Viktor Söderqvist <[email protected]>
Removed the version details and history block from doc files.