mirror of
https://github.com/gogrlx/nats-server.git
synced 2026-04-02 03:38:42 -07:00
fda5bd7ac7
Clients that will be at the ClientProtoInfo protocol level (or above) will now receive an asynchronous INFO protocol when the server they connect to adds a *new* route. This means that when the cluster adds a new server, all clients in the cluster should now be notified of this new addition.
2.2 KiB
2.2 KiB
General
- Auth for queue groups?
- Blacklist or ERR escalation to close connection for auth/permissions
- Protocol updates, MAP, MPUB, etc
- Multiple listen endpoints
- Websocket / HTTP2 strategy
- T series reservations
- _SYS. server events?
- No downtime restart
- Signal based reload of configuration
- brew, apt-get, rpm, chocately (windows)
- IOVec pools and writev for high fanout?
- Modify cluster support for single message across routes between pub/sub and d-queue
- Memory limits/warnings?
- Limit number of subscriptions a client can have, total memory usage etc.
- Multi-tenant accounts with isolation of subject space
- Pedantic state
- _SYS.> reserved for server events?
- Listen configure key vs addr and port
- Add ENV and variable support to dconf? ucl?
- Buffer pools/sync pools?
- Multiple Authorization / Access
- Write dynamic socket buffer sizes
- Read dynamic socket buffer sizes
- Info updates contain other implicit route servers
- Sublist better at high concurrency, cache uses writelock always currently
- Switch to 1.4/1.5 and use maps vs hashmaps in sublist
- NewSource on Rand to lower lock contention on QueueSubs, or redesign!
- Default sort by cid on connz
- Track last activity time per connection?
- Add total connections to varz so we won't miss spikes, etc.
- Add starttime and uptime to connz list.
- Gossip Protocol for discovery for clustering
- Add in HTTP requests to varz?
- Add favico and help link for monitoring?
- Better user/pass support using bcrypt etc.
- SSL/TLS support
- Add support for / to point to varz, connz, etc..
- Support sort options for /connz via nats-top
- Dropped message statistics (slow consumers)
- Add current time to each monitoring endpoint
- varz uptime do days and only integer secs
- Place version in varz (same info sent to clients)
- Place server ID/UUID in varz
- nats-top equivalent, utils
- Connz report routes (/routez)
- Docker
- Remove reliance on
ps - Syslog support
- Client support for language and version
- Fix benchmarks on linux
- Daemon mode? Won't fix