optimize: RM TM startup connect server fail fast

[jsbxyyx]

• I have registered the PR changes.

Ⅰ. Describe what this PR did

fixes: #5484

Ⅱ. Does this pull request fix one issue?

Ⅲ. Why don't you add test cases (unit test/integration test)?

Ⅳ. Describe how to verify it

Ⅴ. Special notes for reviews

[codecov]

Codecov Report

Merging #6004 (e549797) into 2.x (501d431) will decrease coverage by 0.12%.
Report is 1 commits behind head on 2.x.
The diff coverage is 31.88%.

Additional details and impacted files

@@             Coverage Diff              @@
##                2.x    #6004      +/-   ##
============================================
- Coverage     48.98%   48.87%   -0.12%     
+ Complexity     4782     4765      -17     
============================================
  Files           913      913              
  Lines         31697    31750      +53     
  Branches       3823     3835      +12     
============================================
- Hits          15528    15518      -10     
- Misses        14632    14689      +57     
- Partials       1537     1543       +6     

Files	Coverage Δ
...c/main/java/io/seata/common/ConfigurationKeys.java	0.00% <ø> (ø)
...n/src/main/java/io/seata/common/DefaultValues.java	0.00% <ø> (ø)
...io/seata/core/rpc/netty/AbstractNettyRemoting.java	14.28% <ø> (ø)
...n/java/io/seata/core/rpc/netty/ChannelManager.java	0.00% <ø> (ø)
...ava/io/seata/core/rpc/netty/NettyClientConfig.java	44.77% <ø> (ø)
...io/seata/core/rpc/netty/TmNettyRemotingClient.java	55.00% <100.00%> (+0.45%)	⬆️
...va/io/seata/integration/tx/api/util/DubboUtil.java	0.00% <0.00%> (ø)
...io/seata/core/rpc/netty/RmNettyRemotingClient.java	37.31% <0.00%> (-0.57%)	⬇️
...ta/core/rpc/netty/AbstractNettyRemotingClient.java	18.26% <16.66%> (-0.45%)	⬇️
...rc/main/java/io/seata/common/util/StringUtils.java	40.37% <0.00%> (-4.12%)	⬇️
... and 1 more

... and 5 files with indirect coverage changes

[funky-eyes]

LGTM

[laywin]

1. may be fast fail is there are no available server from registry throw exception when start up, not there are available server but some can't connect. the check process can keep the available server list clear.
2. In addition, i think the available server map can hook into registry, when the server not online, available server map can first time get the notify message not depend on 10s to polling.

[slievrly]

1. may be fast fail is there are no available server from registry throw exception when start up, not there are available server but some can't connect. the check process can keep the available server list clear.
2. In addition, i think the available server map can hook into registry, when the server not online, available server map can first time get the notify message not depend on 10s to polling.

The server online/offline notifications in the registry center are based on a subscription mechanism. The "10s" mentioned here refers to the interval for checking the availability of servers. The registry center list displays the nodes that are deemed normal, but it doesn't guarantee that the clients can necessarily connect to them.

[laywin]

1. may be fast fail is there are no available server from registry throw exception when start up, not there are available server but some can't connect. the check process can keep the available server list clear.
2. In addition, i think the available server map can hook into registry, when the server not online, available server map can first time get the notify message not depend on 10s to polling.

The server online/offline notifications in the registry center are based on a subscription mechanism. The "10s" mentioned here refers to the interval for checking the availability of servers. The registry center list displays the nodes that are deemed normal, but it doesn't guarantee that the clients can necessarily connect to them.

Yes, I know the current situation. What I mean is that the list after check can also be referenced in the process of subscription monitoring to determine whether the node can be deleted at the first time when it goes offline, and 10s check is still in place.

是的，我知道现在的情况，我的意思是check 后的列表一样可以引用到订阅监听的过程中去，用来判断当节点下线时，能够第一时间删除掉， 10s check还在.

[codecov-commenter]

Codecov Report

Attention: 43 lines in your changes are missing coverage. Please review.

Comparison is base (6011c68) 49.75% compared to head (0893fca) 49.96%.

Additional details and impacted files

@@             Coverage Diff              @@
##                2.x    #6004      +/-   ##
============================================
+ Coverage     49.75%   49.96%   +0.21%     
- Complexity     4885     4932      +47     
============================================
  Files           915      915              
  Lines         31885    31927      +42     
  Branches       3853     3855       +2     
============================================
+ Hits          15863    15952      +89     
+ Misses        14470    14411      -59     
- Partials       1552     1564      +12     

Files	Coverage Δ
...c/main/java/io/seata/common/ConfigurationKeys.java	0.00% <ø> (ø)
...n/src/main/java/io/seata/common/DefaultValues.java	0.00% <ø> (ø)
...io/seata/core/rpc/netty/AbstractNettyRemoting.java	12.98% <ø> (-1.30%)	⬇️
...ava/io/seata/core/rpc/netty/NettyClientConfig.java	49.25% <ø> (+4.47%)	⬆️
...io/seata/core/rpc/netty/TmNettyRemotingClient.java	71.00% <100.00%> (+16.45%)	⬆️
...io/seata/core/rpc/netty/RmNettyRemotingClient.java	51.49% <25.00%> (+13.61%)	⬆️
...ta/core/rpc/netty/AbstractNettyRemotingClient.java	20.19% <16.66%> (+1.47%)	⬆️
...eata/core/rpc/netty/NettyClientChannelManager.java	62.17% <55.26%> (-0.33%)	⬇️
...rc/main/java/io/seata/common/util/StringUtils.java	46.00% <0.00%> (-2.59%)	⬇️

... and 11 files with indirect coverage changes

[funky-eyes]

LGTM