swift上传大文件出错

发布于 2021-11-29 08:06:12 字数 4283 浏览 923 评论 9

@zhouxingxing 你好，看了你对swift的研究，受益匪浅，不过我这边出现了个奇怪的问题，想跟你请教下：
环境是1proxy、2node，用的是回环文件。上传小文件没问题，但上传1G+的大文件就报错。这是日志：

客户端：

Object PUT failed: http://10.100.0.16:18888/v1/AUTH_test/myfiles/test/aa.data 503 Service Unavailable [first 60 chars of response] <html><h1>Service Unavailable</h1><p>The server is currently

proxy：
Sep 22 13:03:45 localhost proxy-server ERROR with Object server 10.100.0.15:10000/sdb1 re: Trying to write to /v1/AUTH_test/myfiles/test/aa.data: ChunkWriteTimeout (10s)

node1：
Sep 22 13:02:51 localhost account-replicator Beginning replication run
Sep 22 13:03:01 localhost account-replicator ERROR reading HTTP response from {'replication_port': 10002, 'zone': 2, 'weight': 100.0, 'ip': '10.100.0.15', 'region': 1, 'port': 10002, 'replication_ip': '10.100.0.15', 'meta': u'', 'device': 'sdb1', 'id': 1}: Timeout (10s)
Sep 22 13:03:01 localhost account-replicator Replication run OVER
Sep 22 13:03:01 localhost account-replicator Attempted to replicate 1 dbs in 10.00580 seconds (0.09994/s)
Sep 22 13:03:01 localhost account-replicator Removed 0 dbs
Sep 22 13:03:01 localhost account-replicator 0 successes, 1 failures
Sep 22 13:03:01 localhost account-replicator no_change:0 ts_repl:0 diff:0 rsync:0 diff_capped:0 hashmatch:0 empty:0
Sep 22 13:03:02 localhost object-auditor Begin object audit "forever" mode (ALL)
Sep 22 13:03:02 localhost object-auditor Object audit (ALL) "forever" mode completed: 0.10s. Total quarantined: 0, Total errors: 0, Total files/sec: 38.74 , Total bytes/sec: 174472.14, Auditing time: 0.06, Rate: 0.60
Sep 22 13:03:03 localhost container-replicator Beginning replication run
Sep 22 13:03:03 localhost container-replicator Replication run OVER

node2：

Sep 22 13:02:59 localhost container-replicator Attempted to replicate 2 dbs in 0.00945 seconds (211.72436/s)
Sep 22 13:02:59 localhost container-replicator Removed 0 dbs
Sep 22 13:02:59 localhost container-replicator 2 successes, 0 failures
Sep 22 13:02:59 localhost container-replicator no_change:2 ts_repl:0 diff:0 rsync:0 diff_capped:0 hashmatch:0 empty:0
Sep 22 13:02:59 localhost object-auditor Begin object audit "forever" mode (ZBF)
Sep 22 13:02:59 localhost object-auditor Object audit (ZBF) "forever" mode completed: 0.06s. Total quarantined: 0, Total errors: 0, Total files/sec: 64.91 , Total bytes/sec: 0.00, Auditing time: 0.03, Rate: 0.53
Sep 22 13:02:59 localhost object-auditor Begin object audit "forever" mode (ALL)
Sep 22 13:02:59 localhost object-auditor Object audit (ALL) "forever" mode completed: 0.10s. Total quarantined: 0, Total errors: 0, Total files/sec: 39.19 , Total bytes/sec: 176490.90, Auditing time: 0.07, Rate: 0.72

Sep 22 13:02:59 localhost account-server ERROR __call__ error with REPLICATE /sdb1/122360/777e3b96a7936fb783a5ecf8e7d1c95d : LockTimeout (10s) /swift/sdb1/accounts/122360/95d/777e3b96a7936fb783a5ecf8e7d1c95d/.lock
Sep 22 13:02:59 localhost account-replicator Beginning replication run
Sep 22 13:02:59 localhost account-replicator Replication run OVER
Sep 22 13:02:59 localhost account-replicator Attempted to replicate 1 dbs in 6.49791 seconds (0.15390/s)
Sep 22 13:02:59 localhost account-replicator Removed 0 dbs
Sep 22 13:02:59 localhost account-replicator 1 successes, 0 failures
Sep 22 13:02:59 localhost account-replicator no_change:1 ts_repl:0 diff:0 rsync:0 diff_capped:0 hashmatch:0 empty:0
Sep 22 13:02:59 localhost account-server 10.100.0.13 - - [22/Sep/2013:05:02:59 +0000] "PUT /sdb1/122360/AUTH_test/myfiles" 201 - "-" "-" "container-updater 16242" 31.2634 ""
Sep 22 13:02:59 localhost object-replicator 549/549 (100.00%) partitions replicated in 70.73s (7.76/sec, 0s remaining)
Sep 22 13:02:59 localhost object-replicator 549 suffixes checked - 0.00% hashed, 0.00% synced
Sep 22 13:02:59 localhost object-replicator Partition times: max 34.5089s, min 0.0019s, med 0.0024s
Sep 22 13:02:59 localhost object-replicator Object replication complete. (1.18 minutes)
感觉是node2出问题了，好像有文件被锁住了，但找不解决办法......