开发某个项目过程中，就需求，搭建了一套测试环境。很快完成！
后来代码中加入了许多新功能，会涉及到反复重启，然后就发现了启动特别慢。这给原本功能就不多的应用增添了许多的负担。
我决定改变这一切！找到启动缓慢的根源，加快启动速度！
思路展开:
1. 启动慢，先看日志，有没有什么异常？
2. 看下是不是gc 有问题，是否占用很长时间？
3. 日志卡住，是否是存在网络请求不通情况？
4. 抓包查看卡住的时候，应用都做什么？
5. 把线程堆栈打印出来，线程都在做什么？把内存dump 出来瞅瞅吧？
6. 实在不行，本地debug不行，换一台测试机试试？
7. 实在不行，远程debug吧？
8. 定位到点后，再来思考解决好的解决方案？
9. 除了解决这个问题，我们还能思考点什么？
让我按思路一个个来看解决过程！
1. 启动慢，先看日志，有没有什么异常？
这个基本上很快就有答案了，看日志，等待期间没有问题！
但是，我们往往会停留在这个阶段较长一段时间，因为我们总想通过其他方式，打印点什么出来！
所以，我们可以会改日志级别，try... catch... 等等一顿操作，然后，才说算了吧，我们想想别的招！我个人还是觉得这种思路是ok的！
2. 看下是不是gc 有问题，是否占用很长时间？
这里可能就会涉及，之前并没有打印gc日志，然后还得回去在启动参数里加上gc输出！姿势如下:
-Xloggc:/opt/logs/a_gc.log -XX:+PrintGCDetails -XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=/opt/logs/ -XX:+UseG1GC -XX:MaxGCPauseMillis=100 -XX:+ParallelRefProcEnabled -XX:ErrorFile=/run/dump/hs_err_pid%p.lo3g -XX:HeapDumpPath=/run/dump
打开后，你看到的gc 日志可能是这样的。 你想从里面发现点什么，好像频繁没有 FullGC 之类的操作，然后又悻悻而归！
10.224: [GC concurrent-root-region-scan-start] 10.249: [GC concurrent-root-region-scan-end, 0.0252691 secs] 10.249: [GC concurrent-mark-start] 10.250: [GC concurrent-mark-end, 0.0005479 secs] 10.250: [GC remark 10.251: [Finalize Marking, 0.0001862 secs] 10.251: [GC ref-proc, 0.0026452 secs] 10.253: [Unloading, 0.0125200 secs], 0.0156381 secs] [Times: user=0.06 sys=0.00, real=0.02 secs] 10.266: [GC cleanup 33M->33M(1024M), 0.0012152 secs] [Times: user=0.00 sys=0.00, real=0.00 secs] 13.768: [GC pause (G1 Evacuation Pause) (young), 0.0411037 secs] [Parallel Time: 34.5 ms, GC Workers: 4] [GC Worker Start (ms): Min: 13768.7, Avg: 13769.8, Max: 13773.1, Diff: 4.4] [Ext Root Scanning (ms): Min: 0.0, Avg: 1.8, Max: 4.1, Diff: 4.1, Sum: 7.3] [Update RS (ms): Min: 0.0, Avg: 0.0, Max: 0.0, Diff: 0.0, Sum: 0.0] [Processed Buffers: Min: 0, Avg: 0.0, Max: 0, Diff: 0, Sum: 0] [Scan RS (ms): Min: 0.0, Avg: 0.1, Max: 0.2, Diff: 0.2, Sum: 0.5] [Code Root Scanning (ms): Min: 0.0, Avg: 3.1, Max: 8.8, Diff: 8.8, Sum: 12.5] [Object Copy (ms): Min: 21.1, Avg: 28.0, Max: 30.5, Diff: 9.4, Sum: 111.9] [Termination (ms): Min: 0.0, Avg: 0.1, Max: 0.1, Diff: 0.1, Sum: 0.2] [Termination Attempts: Min: 1, Avg: 12.2, Max: 22, Diff: 21, Sum: 49] [GC Worker Other (ms): Min: 0.0, Avg: 0.0, Max: 0.0, Diff: 0.0, Sum: 0.1] [GC Worker Total (ms): Min: 29.9, Avg: 33.2, Max: 34.3, Diff: 4.4, Sum: 132.6] [GC Worker End (ms): Min: 13803.0, Avg: 13803.0, Max: 13803.0, Diff: 0.0] [Code Root Fixup: 0.4 ms] [Code Root Purge: 0.0 ms] [Clear CT: 0.2 ms] [Other: 5.9 ms] [Choose CSet: 0.0 ms] [Ref Proc: 4.6 ms] [Ref Enq: 0.1 ms] [Redirty Cards: 0.1 ms] [Humongous Register: 0.1 ms] [Humongous Reclaim: 0.0 ms] [Free CSet: 0.9 ms] [Eden: 544.0M(544.0M)->0.0B(573.0M) Survivors: 31.0M->41.0M Heap: 575.0M(1024.0M)->41.0M(1024.0M)] [Times: user=0.13 sys=0.02, real=0.04 secs] 17.178: [GC pause (G1 Evacuation Pause) (young), 0.0648366 secs] [Parallel Time: 59.7 ms, GC Workers: 4] [GC Worker Start (ms): Min: 17177.7, Avg: 17181.1, Max: 17184.4, Diff: 6.6] [Ext Root Scanning (ms): Min: 0.0, Avg: 3.9, Max: 12.7, Diff: 12.7, Sum: 15.6] [Update RS (ms): Min: 0.0, Avg: 0.0, Max: 0.0, Diff: 0.0, Sum: 0.0] [Processed Buffers: Min: 0, Avg: 0.0, Max: 0, Diff: 0, Sum: 0] [Scan RS (ms): Min: 0.0, Avg: 0.1, Max: 0.3, Diff: 0.2, Sum: 0.6] [Code Root Scanning (ms): Min: 0.0, Avg: 6.5, Max: 19.4, Diff: 19.4, Sum: 26.0] [Object Copy (ms): Min: 34.9, Avg: 45.5, Max: 51.2, Diff: 16.3, Sum: 182.1] [Termination (ms): Min: 0.0, Avg: 0.1, Max: 0.1, Diff: 0.1, Sum: 0.3] [Termination Attempts: Min: 1, Avg: 16.5, Max: 24, Diff: 23, Sum: 66] [GC Worker Other (ms): Min: 0.0, Avg: 0.1, Max: 0.1, Diff: 0.1, Sum: 0.2] [GC Worker Total (ms): Min: 52.9, Avg: 56.2, Max: 59.5, Diff: 6.6, Sum: 224.8] [GC Worker End (ms): Min: 17237.3, Avg: 17237.3, Max: 17237.3, Diff: 0.0] [Code Root Fixup: 0.5 ms] 20.648: [GC pause (G1 Evacuation Pause) (young), 0.0653569 secs] [Parallel Time: 55.9 ms, GC Workers: 4] [GC Worker Start (ms): Min: 20648.5, Avg: 20650.4, Max: 20652.3, Diff: 3.7] [Ext Root Scanning (ms): Min: 0.1, Avg: 2.9, Max: 9.4, Diff: 9.4, Sum: 11.7] [Update RS (ms): Min: 0.0, Avg: 0.2, Max: 0.5, Diff: 0.5, Sum: 0.7] [Processed Buffers: Min: 0, Avg: 2.5, Max: 7, Diff: 7, Sum: 10] [Scan RS (ms): Min: 0.0, Avg: 0.3, Max: 0.6, Diff: 0.6, Sum: 1.3] [Code Root Scanning (ms): Min: 0.0, Avg: 3.2, Max: 10.8, Diff: 10.8, Sum: 12.9] [Object Copy (ms): Min: 39.4, Avg: 46.4, Max: 52.1, Diff: 12.8, Sum: 185.5] [Termination (ms): Min: 0.0, Avg: 0.9, Max: 1.5, Diff: 1.5, Sum: 3.7] [Termination Attempts: Min: 1, Avg: 1.0, Max: 1, Diff: 0, Sum: 4] [GC Worker Other (ms): Min: 0.0, Avg: 0.0, Max: 0.0, Diff: 0.0, Sum: 0.0] [GC Worker Total (ms): Min: 52.1, Avg: 54.0, Max: 55.8, Diff: 3.7, Sum: 215.8] [GC Worker End (ms): Min: 20704.4, Avg: 20704.4, Max: 20704.4, Diff: 0.0] [Code Root Fixup: 0.2 ms] [Code Root Purge: 0.0 ms] [Clear CT: 0.1 ms] [Other: 9.2 ms] [Choose CSet: 0.0 ms] [Ref Proc: 8.5 ms] [Ref Enq: 0.0 ms] [Redirty Cards: 0.2 ms] [Humongous Register: 0.1 ms] [Humongous Reclaim: 0.0 ms] [Free CSet: 0.2 ms] [Eden: 38.0M(8192.0K)->0.0B(610.0M) Survivors: 43.0M->4096.0K Heap: 83.3M(1024.0M)->54.4M(1024.0M)] [Times: user=0.24 sys=0.02, real=0.06 secs] 2826.037: [GC pause (Metadata GC Threshold) (young) (initial-mark), 0.0101504 secs] [Parallel Time: 7.1 ms, GC Workers: 4] [GC Worker Start (ms): Min: 2826037.4, Avg: 2826037.5, Max: 2826037.5, Diff: 0.1] [Ext Root Scanning (ms): Min: 1.7, Avg: 2.0, Max: 2.2, Diff: 0.5, Sum: 7.8] [Update RS (ms): Min: 0.9, Avg: 1.0, Max: 1.0, Diff: 0.1, Sum: 3.8] [Processed Buffers: Min: 7, Avg: 14.2, Max: 25, Diff: 18, Sum: 57] [Scan RS (ms): Min: 0.3, Avg: 0.7, Max: 0.9, Diff: 0.6, Sum: 2.8] [Code Root Scanning (ms): Min: 0.2, Avg: 0.4, Max: 0.8, Diff: 0.6, Sum: 1.5] [Object Copy (ms): Min: 2.7, Avg: 2.9, Max: 3.2, Diff: 0.5, Sum: 11.7] [Termination (ms): Min: 0.0, Avg: 0.0, Max: 0.0, Diff: 0.0, Sum: 0.0] [Termination Attempts: Min: 1, Avg: 1.0, Max: 1, Diff: 0, Sum: 4] [GC Worker Other (ms): Min: 0.0, Avg: 0.0, Max: 0.0, Diff: 0.0, Sum: 0.1] [GC Worker Total (ms): Min: 6.9, Avg: 7.0, Max: 7.0, Diff: 0.1, Sum: 27.8] [GC Worker End (ms): Min: 2826044.4, Avg: 2826044.4, Max: 2826044.4, Diff: 0.0] [Code Root Fixup: 0.0 ms] [Code Root Purge: 0.0 ms] [Clear CT: 0.1 ms] [Other: 2.9 ms] [Choose CSet: 0.0 ms] [Ref Proc: 1.6 ms] [Ref Enq: 0.1 ms] [Redirty Cards: 0.0 ms] [Humongous Register: 0.0 ms] [Humongous Reclaim: 0.0 ms] [Free CSet: 0.5 ms] [Eden: 354.0M(438.0M)->0.0B(603.0M) Survivors: 4096.0K->11.0M Heap: 408.3M(1024.0M)->61.4M(1024.0M)] [Times: user=0.03 sys=0.00, real=0.01 secs] 2826.047: [GC concurrent-root-region-scan-start] 2826.054: [GC concurrent-root-region-scan-end, 0.0066742 secs] 2826.054: [GC concurrent-mark-start] 2826.116: [GC concurrent-mark-end, 0.0616457 secs] 2826.116: [GC remark 2826.116: [Finalize Marking, 0.0002202 secs] 2826.116: [GC ref-proc, 0.0031923 secs] 2826.120: [Unloading, 0.0131458 secs], 0.0169584 secs] [Times: user=0.05 sys=0.00, real=0.01 secs] 2826.134: [GC cleanup 62M->62M(1024M), 0.0010641 secs] [Times: user=0.00 sys=0.00, real=0.00 secs]
不管怎么样，你知道内存不是问题了。不过一般地，在启动时就存在内存问题的应用可能确实不多！
3. 日志卡住，是否是存在网络请求不通情况？
一般来说，我们的应用，每做一些关键操作时，都会有相应的日志输出。所以，当你日志卡住的时候，应该就是哪里出现了问题了！
而在排除了中间出现full gc 卡顿的问题后，我们可以认为可能是网络出现问题了。
查询网络问题必备的工具: tcpdump, lsof ...
用法也很简单，我们可以在完全没有任何信息的前提下，进行网络抓包，从而给排查问题一点提示:
tcpdump -iany -XX # 这样就可以看到应用的网络io情况了, 如果量太大可以先把 -XX 选项去除，只看来往情况:
大概结果如下:
16:54:25.770463 IP 10.5.3.215.63773 > zt-d-xx.ssh: Flags [.], ack 125340, win 552, length 0 16:54:25.770478 IP zt-d-xx.ssh > 10.5.3.215.63773: Flags [P.], seq 127152:127332, ack 1, win 561, length 180 16:54:25.770482 IP 10.1.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 125340, win 552, options [nop,nop,sack 1 {125504:125684}], length 0 16:54:25.770487 IP 10.1.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 125684, win 556, length 0 16:54:25.770554 IP zt-d-xx.ssh > 10.1.2.215.63773: Flags [P.], seq 127332:127864, ack 1, win 561, length 532 16:54:25.770579 IP zt-d-xx.ssh > 10.1.2.215.63773: Flags [P.], seq 127864:127916, ack 1, win 561, length 52 16:54:25.770612 IP zt-d-xx.ssh > 10.1.2.215.63773: Flags [P.], seq 127916:128208, ack 1, win 561, length 292 16:54:25.771813 IP 10.1.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 125684, win 556, options [nop,nop,sack 1 {125864:126044}], length 0 16:54:25.771822 IP zt-d-xx.ssh > 10.1.2.215.63773: Flags [P.], seq 128208:128388, ack 1, win 561, length 180 16:54:25.771837 IP 10.1.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 126044, win 555, length 0 16:54:25.771840 IP 10.1.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 126792, win 552, length 0 16:54:25.771868 IP zt-d-xx.ssh > 10.1.2.215.63773: Flags [P.], seq 128388:128808, ack 1, win 561, length 420 16:54:25.771884 IP zt-d-xx.ssh > 10.1.2.215.63773: Flags [P.], seq 128808:128972, ack 1, win 561, length 164 16:54:25.771891 IP 10.1.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 127152, win 556, length 0 16:54:25.771934 IP zt-d-xx.ssh > 10.1.2.215.63773: Flags [P.], seq 128972:129376, ack 1, win 561, length 404 16:54:25.771959 IP zt-d-xx.ssh > 10.5.2.215.63773: Flags [P.], seq 129376:129556, ack 1, win 561, length 180 16:54:25.771982 IP zt-d-xx.ssh > 10.5.2.215.63773: Flags [P.], seq 129556:129736, ack 1, win 561, length 180 16:54:25.779454 IP 10.5.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 127864, win 553, length 0 16:54:25.779469 IP zt-d-xx.ssh > 10.5.2.215.63773: Flags [P.], seq 129736:129916, ack 1, win 561, length 180 16:54:25.779474 IP 10.5.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 128208, win 552, length 0 16:54:25.779476 IP 10.5.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 128808, win 556, length 0 16:54:25.779478 IP 10.5.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 129376, win 554, length 0 16:54:25.779480 IP 10.5.2.215.63773 > zt-d-xx.ssh: Flags [.], ack 129736, win 553, length 0 16:54:25.779529 IP zt-d-xx.ssh > 10.5.3.215.63773: Flags [P.], seq 129916:130496, ack 1, win 561, length 580 16:54:25.779567 IP zt-d-xx.ssh > 10.5.3.215.63773: Flags [P.], seq 130496:130660, ack 1, win 561, length 164
其中，我们主要关注 Flags 参数. man tcpdump 后可以看到解释:
Tcpflags are some combination of S (SYN), F (FIN), P (PUSH), R (RST), U (URG), W (ECN CWR), E (ECN-Echo) or '.' (ACK)
详细说明就是:
# SYN 表示建立连接;
# FIN 表示关闭连接;
# ACK 表示响应;
# PSH 表示有 DATA数据传输;
# RST 表示连接重置;
# URG 表示紧急停止位;
所以，我们在这里需要关注这么多吗？ 其实不需要的，我觉得咱们可以先看下是否有 RST 连接重置的情况，如果存在，则说明这个网络是不通的，基本定位网络问题！~ 如下抓包:
17:30:41.635937 IP zt-d-xx.43062 > 172.1.0.17.http: Flags [S], seq 3503889751, win 29200, options [mss 1460,sackOK,TS val 1904549734 ecr 0,nop,wscale 7], length 0 17:30:41.636084 IP 172.1.0.17.http > zt-d-xx.43062: Flags [R.], seq 0, ack 3503889752, win 0, length 0
另外，要查看下是不是某个网络请求 只有 [S] 信号，没有 [P] 的信号, 则说明只有一端在建立连接，另一边则无响应，网络问题定位！如下请求一个不存在的地址:
17:13:57.744349 IP zt-d-xx.37604 > 172-1-1-2.lightspeed.hstntx.sbcglobal.net.http: Flags [S], seq 3312915935, win 29200, options [mss 1460,sackOK,TS val 1903545842 ecr 0,nop,wscale 7], length 0 17:13:58.745584 IP zt-d-xx.37604 > 172-1-1-2.lightspeed.hstntx.sbcglobal.net.http: Flags [S], seq 3312915935, win 29200, options [mss 1460,sackOK,TS val 1903546844 ecr 0,nop,wscale 7], length 0 17:14:00.749570 IP zt-d-xx.37604 > 172-1-1-2.lightspeed.hstntx.sbcglobal.net.http: Flags [S], seq 3312915935, win 29200, options [mss 1460,sackOK,TS val 1903548848 ecr 0,nop,wscale 7], length 0 17:14:04.757571 IP zt-d-xx.37604 > 172-1-1-2.lightspeed.hstntx.sbcglobal.net.http: Flags [S], seq 3312915935, win 29200, options [mss 1460,sackOK,TS val 1903552856 ecr 0,nop,wscale 7], length 0 17:14:12.765568 IP zt-d-xx.37604 > 172-1-1-2.lightspeed.hstntx.sbcglobal.net.http: Flags [S], seq 3312915935, win 29200, options [mss 1460,sackOK,TS val 1903560864 ecr 0,nop,wscale 7], length 0 17:14:28.797584 IP zt-d-xx.37604 > 172-1-1-2.lightspeed.hstntx.sbcglobal.net.http: Flags [S], seq 3312915935, win 29200, options [mss 1460,sackOK,TS val 1903576896 ecr 0,nop,wscale 7], length 0 17:15:00.861591 IP zt-d-xx.37604 > 172-1-1-2.lightspeed.hstntx.sbcglobal.net.http: Flags [S], seq 3312915935, win 29200, options [mss 1460,sackOK,TS val 1903608960 ecr 0,nop,wscale 7], length 0
发了n个 [S] 包，都没有响应; 说明整个网络都是不通的;
通过这个排查，我们一般排除网络问题！
4. 抓包查看卡住的时候，应用都做什么？
这个点其实可以早点做，使用 top 查看内存，cpu 占用情况！
使用 vmstat 查看系统虚拟内存状态，也可以报告关于进程、内存、I/O等系统整体运行状态！
使用 lsof -p <pid> 查看打开的文件情况！应该可以看到一些不一样的东西，比如你会看到一些你不知道被打开的文件，你会看到一些你不知道的连接请求！
5. 把线程堆栈打印出来，线程都在做什么？把内存dump 出来瞅瞅吧？
如果还是没有发现什么，那我们还得回来。看看线程卡在了哪里！
使用 jstack 把线程堆栈打印出来，使用 jmap 把内存dump 出来，分析:
jstack <pid> | less # 查看线程信息，是否存在死锁 jmap -dump:format=b,file=/tmp/dumpid.dump <pid> # 查看堆信息，是否需要特别的对象
6. 实在不行，本地debug不行，换一台测试机试试？
对于本地debug, 我只能说，一般环境都不通的，而且本地一般也很复现场景！
换台机器测试的目的，其实是想确认问题是否在所有机器上复现，因为如果是和机器本身相关的问题，往往是很难排查的。（我说的没有相关经验的同学）
所以，找一台另外机器，代码跑起来，如果其他地方正常，则该问题是机器相关的。
那么机器相关的bug又该怎么办呢？其实，还得具体问题，具体分析！
不过幸好，我们还有 remote debug 功能，直接连接上去就可以调试了。问题自然也就清晰起来！
7. 实在不行，远程debug吧？
remote debug 请查看上一篇文章: 排障利器之远程调试与监控 --jmx & remote debug
连接上之后，先大概猜测可能会出问题的地方，进行断点。判断依据，应该是之前的排查结果，线程情况等等！
进行大概估计后，在可能的地方进行单步，卡住的地方，其中必然出现了问题。只需再进入内部重复刚才的做法即可！
优点是这样调试下来，必定能找到问题出现的地方。缺点是：可能需要反复重现问题（可能就是反复重启）。但是这样显得没有技术含量，还有就是在问题难以复现的场景，很难抓住机会解决问题。
所以，经验真的很重要！
卡住时的堆栈如下:
"main@1" prio=5 tid=0x1 nid=NA runnable java.lang.Thread.State: RUNNABLE at java.io.FileInputStream.readBytes(FileInputStream.java:-1) at java.io.FileInputStream.read(FileInputStream.java:255) at sun.security.provider.SeedGenerator$URLSeedGenerator.getSeedBytes(SeedGenerator.java:539) at sun.security.provider.SeedGenerator.generateSeed(SeedGenerator.java:144) at sun.security.provider.SecureRandom$SeederHolder.<clinit>(SecureRandom.java:203) at sun.security.provider.SecureRandom.engineNextBytes(SecureRandom.java:221) - locked <0x1af3> (a sun.security.provider.SecureRandom) at java.security.SecureRandom.nextBytes(SecureRandom.java:468) at java.security.SecureRandom.next(SecureRandom.java:491) at java.util.Random.nextLong(Random.java:424) at com.taobao.notify.utils.StrongRandom.<init>(StrongRandom.java:45) at com.taobao.notify.utils.UniqId.<init>(UniqId.java:38) at com.taobao.notify.utils.UniqId.<clinit>(UniqId.java:36) at com.taobao.notify.client.impl.DefaultNotifyClient.<init>(DefaultNotifyClient.java:91) at com.taobao.notify.client.impl.DefaultNotifyClient.<init>(DefaultNotifyClient.java:97) at com.taobao.notify.client.NotifyClientFactory.<clinit>(NotifyClientFactory.java:13) at com.taobao.notify.remotingclient.DefaultNotifyManager.<init>(DefaultNotifyManager.java:38) at com.taobao.notify.remotingclient.NotifyManagerBean.init(NotifyManagerBean.java:90) - locked <0x1a78> (a com.alipay.common.event.tbnotify.adapter.UniformEventPublisherAdapter) at com.alipay.common.event.tbnotify.client.NotifyManagerBeanEx.init(NotifyManagerBeanEx.java:64) at com.alipay.common.event.tbnotify.adapter.EventSendFacadeTBNotifyImpl.init(EventSendFacadeTBNotifyImpl.java:99) at com.alipay.common.event.tbnotify.client.NotifyManagerBeanEx.init(NotifyManagerBeanEx.java:84) at com.alipay.common.event.tbnotify.adapter.UniformEventPublisherAdapter.init(UniformEventPublisherAdapter.java:80) at com.alipay.boot.dms.client.util.InitializeUtil.initializePublisher(InitializeUtil.java:14) at com.alipay.boot.dms.spring.factory.DmsPublisherFactoryBean.afterPropertiesSet(DmsPublisherFactoryBean.java:59) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.invokeInitMethods(AbstractAutowireCapableBeanFactory.java:1642) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.initializeBean(AbstractAutowireCapableBeanFactory.java:1579) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.doCreateBean(AbstractAutowireCapableBeanFactory.java:553) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.createBean(AbstractAutowireCapableBeanFactory.java:482) at org.springframework.beans.factory.support.AbstractBeanFactory$1.getObject(AbstractBeanFactory.java:306) at org.springframework.beans.factory.support.DefaultSingletonBeanRegistry.getSingleton(DefaultSingletonBeanRegistry.java:230) - locked <0x1af4> (a java.util.concurrent.ConcurrentHashMap) at org.springframework.beans.factory.support.AbstractBeanFactory.doGetBean(AbstractBeanFactory.java:302) at org.springframework.beans.factory.support.AbstractBeanFactory.getBean(AbstractBeanFactory.java:197) at org.springframework.beans.factory.support.BeanDefinitionValueResolver.resolveReference(BeanDefinitionValueResolver.java:351) at org.springframework.beans.factory.support.BeanDefinitionValueResolver.resolveValueIfNecessary(BeanDefinitionValueResolver.java:108) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.applyPropertyValues(AbstractAutowireCapableBeanFactory.java:1486) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.populateBean(AbstractAutowireCapableBeanFactory.java:1231) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.doCreateBean(AbstractAutowireCapableBeanFactory.java:551) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.createBean(AbstractAutowireCapableBeanFactory.java:482) at org.springframework.beans.factory.support.AbstractBeanFactory$1.getObject(AbstractBeanFactory.java:306) at org.springframework.beans.factory.support.DefaultSingletonBeanRegistry.getSingleton(DefaultSingletonBeanRegistry.java:230) at org.springframework.beans.factory.support.AbstractBeanFactory.doGetBean(AbstractBeanFactory.java:302) at org.springframework.beans.factory.support.AbstractBeanFactory.getBean(AbstractBeanFactory.java:202) at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor.autowireResource(CommonAnnotationBeanPostProcessor.java:522) at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor.getResource(CommonAnnotationBeanPostProcessor.java:496) at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor$ResourceElement.getResourceToInject(CommonAnnotationBeanPostProcessor.java:627) at org.springframework.beans.factory.annotation.InjectionMetadata$InjectedElement.inject(InjectionMetadata.java:169) at org.springframework.beans.factory.annotation.InjectionMetadata.inject(InjectionMetadata.java:88) at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor.postProcessPropertyValues(CommonAnnotationBeanPostProcessor.java:318) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.populateBean(AbstractAutowireCapableBeanFactory.java:1219) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.doCreateBean(AbstractAutowireCapableBeanFactory.java:551) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.createBean(AbstractAutowireCapableBeanFactory.java:482) at org.springframework.beans.factory.support.AbstractBeanFactory$1.getObject(AbstractBeanFactory.java:306) at org.springframework.beans.factory.support.DefaultSingletonBeanRegistry.getSingleton(DefaultSingletonBeanRegistry.java:230) at org.springframework.beans.factory.support.AbstractBeanFactory.doGetBean(AbstractBeanFactory.java:302) at org.springframework.beans.factory.support.AbstractBeanFactory.getBean(AbstractBeanFactory.java:202) at org.springframework.beans.factory.config.DependencyDescriptor.resolveCandidate(DependencyDescriptor.java:207) at org.springframework.beans.factory.support.DefaultListableBeanFactory.doResolveDependency(DefaultListableBeanFactory.java:1131) at org.springframework.beans.factory.support.DefaultListableBeanFactory.resolveDependency(DefaultListableBeanFactory.java:1059) at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor.autowireResource(CommonAnnotationBeanPostProcessor.java:518) at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor.getResource(CommonAnnotationBeanPostProcessor.java:496) at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor$ResourceElement.getResourceToInject(CommonAnnotationBeanPostProcessor.java:627) at org.springframework.beans.factory.annotation.InjectionMetadata$InjectedElement.inject(InjectionMetadata.java:169) at org.springframework.beans.factory.annotation.InjectionMetadata.inject(InjectionMetadata.java:88) at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor.postProcessPropertyValues(CommonAnnotationBeanPostProcessor.java:318) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.populateBean(AbstractAutowireCapableBeanFactory.java:1219) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.doCreateBean(AbstractAutowireCapableBeanFactory.java:551) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.createBean(AbstractAutowireCapableBeanFactory.java:482) at org.springframework.beans.factory.support.AbstractBeanFactory$1.getObject(AbstractBeanFactory.java:306) at org.springframework.beans.factory.support.DefaultSingletonBeanRegistry.getSingleton(DefaultSingletonBeanRegistry.java:230) at org.springframework.beans.factory.support.AbstractBeanFactory.doGetBean(AbstractBeanFactory.java:302) at org.springframework.beans.factory.support.AbstractBeanFactory.getBean(AbstractBeanFactory.java:197) at org.springframework.beans.factory.support.DefaultListableBeanFactory.preInstantiateSingletons(DefaultListableBeanFactory.java:754) at org.springframework.context.support.AbstractApplicationContext.finishBeanFactoryInitialization(AbstractApplicationContext.java:866) at org.springframework.context.support.AbstractApplicationContext.refresh(AbstractApplicationContext.java:542) - locked <0x1af5> (a java.lang.Object) at org.springframework.boot.context.embedded.EmbeddedWebApplicationContext.refresh(EmbeddedWebApplicationContext.java:122) at org.springframework.boot.SpringApplication.refresh(SpringApplication.java:761) at org.springframework.boot.SpringApplication.refreshContext(SpringApplication.java:371) at org.springframework.boot.SpringApplication.run(SpringApplication.java:315) at org.springframework.boot.SpringApplication.run(SpringApplication.java:1186) at org.springframework.boot.SpringApplication.run(SpringApplication.java:1175) at com.xx.SOFABootSpringApplication.main(SOFABootSpringApplication.java:15) at sun.reflect.NativeMethodAccessorImpl.invoke0(NativeMethodAccessorImpl.java:-1) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:498) at org.springframework.boot.loader.MainMethodRunner.run(MainMethodRunner.java:48) at org.springframework.boot.loader.Launcher.launch(Launcher.java:87) at org.springframework.boot.loader.Launcher.launch(Launcher.java:50) at org.springframework.boot.loader.JarLauncher.main(JarLauncher.java:58)
而根据堆栈进行的大概猜想，开始排查，过程如下:
// 初始代码位置: // com.taobao.notify.remotingclient.NotifyManagerBean public synchronized void init() { if (inited.get()) { return; } if (StringUtil.isEmpty(this.name) || StringUtil.isEmpty(this.description)) { throw new RuntimeException("The name and description property is required."); } if (null == notifyManager) { AllNotifyClientProperties properties = new AllNotifyClientProperties(); properties.setCheckMessageTPConfig(checkMessageTPConfig); properties.setMessageTPConfig(messageTPConfig); properties.setMessageProperties(messageProperties); properties.setAntVipConfig(antVipConfig); // 问题出在这里，new 这个对象时，会触发很多的其他类加载，而这里面就包含了有问题的类 notifyManager = new DefaultNotifyManager(groupId, appName, name, description, messageListener, checkMessageListener, 1, enableCrc, properties); notifyManager.subscribeAllMessages(this.subscribeMessages, this.bindingList); /** ������ظ���groupId������ɾ��֮ǰ��publish topic*/ //notifyManager.resetPublishTopics(publishTopics); for (String publishTopic : publishTopics) { notifyManager.addPublishTopic(publishTopic); } // ��ʼ��ʱ����servertag��url�Ķ�Ӧ�б� notifyManager.resetServerTagUrls(); } inited.set(true); } // com.taobao.notify.remotingclient.DefaultNotifyManager 这个构造方法是 ok 的 public DefaultNotifyManager(String groupId, String appName, String name, String description, MessageListener msgListener, CheckMessageListener checkMsgListener, int connCount, boolean enableCrc, AllNotifyClientProperties properties) { initNotifyClient(properties); if (StringUtil.isBlank(groupId)) { throw new IllegalArgumentException("GroupID cannot be blank."); } if (StringTools.containsWhitespace(groupId)) { throw new IllegalArgumentException("GroupID cannot contain the blank character. [" + groupId + "]"); } this.groupId = groupId.trim(); this.appName = StringUtil.trim(appName); if (null == properties) { properties = new AllNotifyClientProperties(); } this.notifyclient.addGroup(this.groupId, this.appName, name, description, msgListener, checkMsgListener, properties.getMessageProperties(), properties.getWaitForConnTime(), enableCrc, properties.getAntVipConfig()); this.notifyclient.setConnectionCount(connCount); } // 但是问题在静态字段的初始化问题 // com.taobao.notify.remotingclient.DefaultNotifyManager public class DefaultNotifyManager implements NotifyManager, DefaultNotifyManagerMBean { //��־ private static final Logger logger = MsgBrokerClientLoggerFactory .getLogger(DefaultNotifyManager.class); // 这看起来像是个单例的 client private NotifyClient notifyclient = NotifyClientFactory .getSingletonNotifyClient(); } // com.taobao.notify.client.NotifyClientFactory public class NotifyClientFactory implements FactoryBean { // 初始化类时，会先初始化一个单例 static NotifyClient notifyClient = new DefaultNotifyClient(); /** * * @return */ public static NotifyClient getSingletonNotifyClient() { return notifyClient; } } // 到此，一切看起来都很美好，让我继续往下看 // com.taobao.notify.client.impl.DefaultNotifyClient public class DefaultNotifyClient implements NotifyClient { private static final String LogPrefix = LoggerPrefix .makeLogPrefix(DefaultNotifyClient.class); static final Logger logger = MsgBrokerClientLoggerFactory .getLogger(DefaultNotifyClient.class); private volatile NotifyClientConfig notifyClientConfig = null; private static final int MAX_TIMES = 3; protected final NotifyGroupManager notifyGroupManager = new NotifyGroupManager(); private final RemotingService remotingService; protected final ClientSubscriptionManager clientSubscriptionManager = new ClientSubscriptionManager(); protected final PublishTopicsManager publishTopicsManager = new PublishTopicsManager(); private final ThreadPoolExecutor asynSendMessageWorkTP; private final LoggingService loggingService = LoggingService .getInstance(); private final ReliableAsynSendManager reliableAsynSendManager; private final MessageProperties reliableMessageProperties = new MessageProperties(); // 问题在于 UniqId 的生成 private final UniqId uniqIdInstance = UniqId.getInstance(); /** * */ public DefaultNotifyClient() { this(new NotifyClientConfig()); } /** * @param notifyClientConfig */ public DefaultNotifyClient(final NotifyClientConfig notifyClientConfig) { if (null == this.notifyClientConfig) { this.notifyClientConfig = notifyClientConfig; } if (notifyClientConfig.isDebug()) { this.remotingService = new IntegratedMockRemotingService( this.clientSubscriptionManager, notifyClientConfig.getDebugRemoteSubscribers(), this.notifyGroupManager, notifyClientConfig.getDebugLocalPort()); logger.warn(LogPrefix + "Note that, this is a MsgbrokerClient in the DEBUG mode."); logger.warn(LogPrefix + "In the DEBUG mode, local port [" + notifyClientConfig.getDebugLocalPort() + "]"); logger.warn(LogPrefix + "In the DEBUG mode, connections [" + notifyClientConfig.getDebugRemoteSubscribers() + "]"); } else { this.remotingService = new DefaultRemotingService( this.notifyClientConfig.getRemotingType(), this.notifyGroupManager, notifyClientConfig); } // ��ֹѹ�� this.reliableMessageProperties.setCompressSize(Integer.MAX_VALUE); this.reliableMessageProperties.setMaxStringMessageSize(Integer.MAX_VALUE); this.asynSendMessageWorkTP = new ManagedThreadPoolExecutor(notifyClientConfig .getAsynSendMessageTPConfig().getCorePoolSize(), notifyClientConfig .getAsynSendMessageTPConfig().getMaxPoolSize(), notifyClientConfig .getAsynSendMessageTPConfig().getKeepAliveTime(), notifyClientConfig .getAsynSendMessageTPConfig().getMaxQueueSize(), "asynSendMsgTP-" + this.hashCode(), new ThreadPoolExecutor.AbortPolicy()); this.reliableAsynSendManager = new ReliableAsynSendManager(this, notifyGroupManager, notifyClientConfig); if (notifyClientConfig.isPreInitializeReliableAsynSendManager()) { this.reliableAsynSendManager.init(); } // DefaultNotifyClientʵ�������ж������Ҫ���䲻ͬ��ID try { ThreadPoolConfig messageTPConfig = notifyClientConfig.getMessageTPConfig(); Lookout.registry().info( Lookout.registry().createId("msgbroker.client.group.infos") .withTag("hashcode", String.valueOf(this.hashCode())) .withTag("corePoolSize", String.valueOf(messageTPConfig.getCorePoolSize())) .withTag("maxPoolSize", String.valueOf(messageTPConfig.getMaxPoolSize())) .withTag("queueSize", String.valueOf(messageTPConfig.getMaxQueueSize())) .withTag("channelSize", String.valueOf(messageTPConfig.getChannelSize())), new Info<Set<String>>() { public Set<String> value() { return notifyGroupManager.getGroupIds(); } }); } catch (Exception e) { logger.warn("register msgbroker.client.group.infos failed, " + e.getMessage()); } } } // UniqId 生成如下 // com.taobao.notify.utils.UniqId public class UniqId { private static final Logger log = MsgBrokerClientLoggerFactory.getLogger(UniqId.class); private static final String LogPrefix = LoggerPrefix.makeLogPrefix(UniqId.class); private static char[] digits = { '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'a', 'b', 'c', 'd', 'e', 'f' }; private static Map<Character, Integer> rDigits = new HashMap<Character, Integer>(16); static { for (int i = 0; i < digits.length; ++i) { rDigits.put(digits[i], i); } } private static UniqId me = new UniqId(); private String hostAddr; // 实际问题在于这里了， StrongRandom 的问题 private Random random = new StrongRandom(); private MessageDigest mHasher; private UniqTimer timer = new UniqTimer(); private ReentrantLock opLock = new ReentrantLock(); private UniqId() { try { InetAddress addr = InetAddress.getLocalHost(); hostAddr = addr.getHostAddress(); } catch (IOException e) { log.error(LogPrefix + "Get HostAddr Error", e); hostAddr = String.valueOf(System.currentTimeMillis()); } if (Util.isBlank(hostAddr) || "127.0.0.1".equals(hostAddr)) { hostAddr = String.valueOf(System.currentTimeMillis()); } if (log.isDebugEnabled()) { log.debug(LogPrefix + "hostAddr is:" + hostAddr); } try { mHasher = MessageDigest.getInstance("MD5"); } catch (NoSuchAlgorithmException nex) { mHasher = null; log.error(LogPrefix + "new MD5 Hasher error", nex); } } /** * ��ȡUniqIDʵ�� * * @return UniqId */ public static UniqId getInstance() { return me; } } // StrongRandom 生成缓慢 // com.taobao.notify.utils.StrongRandom public StrongRandom() { try { // This operation may block on some systems with low entropy. See // this page // for workaround suggestions: // http://docs.codehaus.org.display.JETTY.Connectors+slow+to+startup random = SecureRandom.getInstance(SESSION_ID_RANDOM_ALGORITHM); } catch (NoSuchAlgorithmException e) { try { random = SecureRandom.getInstance(SESSION_ID_RANDOM_ALGORITHM_ALT); weakRandom = false; } catch (NoSuchAlgorithmException e_alt) { random = new Random(); weakRandom = true; } } // 就是这里缓慢了，卡住大约一分钟，且无法调试 random.setSeed(random.nextLong() ^ System.currentTimeMillis() ^ hashCode() ^ Runtime.getRuntime().freeMemory()); }
最后，定位到是 随机数生成有问题了！
8. 定位到点后，再来思考解决好的解决方案？
如上，我们知道了是因为 SecureRandom 生成随机数时出现了问题！
那么，如何解决？我能想到的，就是网上搜索答案了。因为这时候已经没有什么可以参考的了。
搜索内容就是， SecureRandom 生成随机缓慢解决方案？ 然后就有答案了！
当然，你可以咨询遇到过这些问题的前辈们，那样更快速！
原因如下：
java生成随机码时，会使用两个文件：
1. /dev/random ， 随机性高，和真实的物理环境有关，阻塞模式。（本文出现的启动缓慢问题，就出在这个上面）
2. /dev/urandom ，伪随机模式，非阻塞，随机性不如 random。
所以，既然是生成random模式有问题，那么想办法换掉这个模式就ok了。
jdk中，提供相应的替换方法，我们先看 $JAVA_HOME/jre/lib/security/java.security 这个里面有个随机数的配置，其说明如下：
# # Sun Provider SecureRandom seed source. # # Select the primary source of seed data for the "SHA1PRNG" and # "NativePRNG" SecureRandom implementations in the "Sun" provider. # (Other SecureRandom implementations might also use this property.) # # On Unix-like systems (for example, Solaris/Linux/MacOS), the # "NativePRNG" and "SHA1PRNG" implementations obtains seed data from # special device files such as file:/dev/random. # # On Windows systems, specifying the URLs "file:/dev/random" or # "file:/dev/urandom" will enable the native Microsoft CryptoAPI seeding # mechanism for SHA1PRNG. # # By default, an attempt is made to use the entropy gathering device # specified by the "securerandom.source" Security property. If an # exception occurs while accessing the specified URL: # # SHA1PRNG: # the traditional system/thread activity algorithm will be used. # # NativePRNG: # a default value of /dev/random will be used. If neither # are available, the implementation will be disabled. # "file" is the only currently supported protocol type. # # The entropy gathering device can also be specified with the System # property "java.security.egd". For example: # # % java -Djava.security.egd=file:/dev/random MainClass # # Specifying this System property will override the # "securerandom.source" Security property. # # In addition, if "file:/dev/random" or "file:/dev/urandom" is # specified, the "NativePRNG" implementation will be more preferred than # SHA1PRNG in the Sun provider. # securerandom.source=file:/dev/random
即 可以直接改这个文件，也可以通过命令行加参数修改，且命令行优先级更高，所以我们使用命令行修改即可：
-Djava.security.egd=file:/dev/./urandom # 在项目中添加java_opts，里面添加启动参数 urandom
如此，再次验证后，启动正常了。问题done。
9. 除了解决这个问题，我们还能思考点什么？
关于 SecureRandom 为什么慢的原因，可以看下这篇文章:
https://blog.csdn.net.zengdeqing2012/article.details/78133370
老话：说了不一定有机会，但不说一定没机会。