errpt 问题

congrd

昨天小机P520报错如下：864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\nFA723BD9   0922141905 I S topsvcs        DeadMan Switch (DMS) close to trigger\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n864D2CE3   0922141905 P S topsvcs        NIM thread blocked\r\n3C81E43F   0922141905 P U topsvcs        Late in sending heartbeat\r\n\r\n\r\n\r\n---------------------------------------------------------------------------\r\nLABEL:          TS_NIM_ERROR_STUCK_\r\nIDENTIFIER:     864D2CE3\r\n\r\nDate/Time:       Thu Sep 22 14:19:23 BEIST 2005\r\nSequence Number: 443\r\nMachine Id:      00C1199F4C00\r\nNode Id:         p201_host\r\nClass:           S\r\nType:            PERM\r\nResource Name:   topsvcs\r\n\r\nDescription\r\nNIM thread blocked\r\n\r\nProbable Causes\r\nA thread in a Topology Services Network Interface Module (NIM) process\r\nwas blocked\r\nTopology Services NIM process cannot get timely access to CPU\r\n\r\nUser Causes\r\nExcessive memory consumption is causing high memory contention\r\nExcessive disk I/O is causing high memory contention\r\n\r\n        Recommended Actions\r\n        Examine I/O and memory activity on the system\r\n        Reduce load on the system\r\n        Tune virtual memory parameters\r\n        Call IBM Service if problem persists\r\n\r\nFailure Causes\r\nExcessive virtual memory activity prevents NIM from making progress\r\nExcessive disk I/O traffic is interfering with paging I/O\r\n\r\n        Recommended Actions\r\n        Examine I/O and memory activity on the system\r\n        Reduce load on the system\r\n        Tune virtual memory parameters\r\n        Call IBM Service if problem persists\r\n\r\nDetail Data\r\nDETECTING MODULE\r\nrsct,nim_control.C,1.38,5455\r\nERROR ID\r\n6XnGH40fnYA1/.vW05g8fg0...................\r\nREFERENCE CODE\r\n\r\nThread which was blocked\r\nreceive thread\r\nInterval in seconds during which process was blocked\r\n         120\r\nInterface name\r\nen2\r\n\r\np201_host#errpt -aj FA723BD9\r\n---------------------------------------------------------------------------\r\nLABEL:          TS_DMS_WARNING_ST\r\nIDENTIFIER:     FA723BD9\r\n\r\nDate/Time:       Thu Sep 22 14:19:23 BEIST 2005\r\nSequence Number: 437\r\nMachine Id:      00C1199F4C00\r\nNode Id:         p201_host\r\nClass:           S\r\nType:            INFO\r\nResource Name:   topsvcs\r\n\r\nDescription\r\nDeadMan Switch (DMS) close to trigger\r\n\r\nProbable Causes\r\nTopology Services daemon cannot get timely access to CPU\r\n\r\nUser Causes\r\nExcessive I/O load is causing high I/O interrupt traffic\r\nExcessive memory consumption is causing high memory contention\r\n\r\n        Recommended Actions\r\n        Reduce application load on the system\r\n        Change (relax) Topology Services tunable parameters\r\n        Call IBM Service if problem persists\r\n\r\nFailure Causes\r\nProblem in Operating System prevents processes from running\r\nExcessive I/O interrupt traffic prevents processes from running\r\nExcessive virtual memory activity prevents Topology Services from making progress\r\nThe system clock was set forward\r\n\r\n        Recommended Actions\r\n        Examine I/O and memory activity on the system\r\n        Reduce load on the system\r\n        Change (relax) Topology Services tunable parameters\r\n        Call IBM Service if problem persists\r\n\r\nDetail Data\r\nDETECTING MODULE\r\nrsct,haDMS_kex.c   1.3.1.5,607\r\nERROR ID\r\n\r\nREFERENCE CODE\r\n\r\nTime remaining until DMS triggers (in msec)\r\n           0\r\nDMS trigger interval (in msec)\r\n       20000\r\n\r\n\r\np201_host#errpt -aj 3C81E43F\r\n---------------------------------------------------------------------------\r\nLABEL:          TS_LATEHB_PE\r\nIDENTIFIER:     3C81E43F\r\n\r\nDate/Time:       Thu Sep 22 14:19:23 BEIST 2005\r\nSequence Number: 430\r\nMachine Id:      00C1199F4C00\r\nNode Id:         p201_host\r\nClass:           U\r\nType:            PERF\r\nResource Name:   topsvcs\r\nResource Class:  NONE\r\nResource Type:   NONE\r\nLocation:\r\nVPD:\r\n\r\nDescription\r\nLate in sending heartbeat\r\n\r\nProbable Causes\r\nHeavy CPU load\r\nSevere physical memory shortage\r\nHeavy I/O activities\r\n\r\nFailure Causes\r\nDaemon can not get required system resource\r\n\r\n        Recommended Actions\r\n        Reduce the system load\r\n\r\nDetail Data\r\nDETECTING MODULE\r\nrsct,bootstrp.C,1.184,4520\r\nERROR ID\r\n6zESUw.fnYA1/MaI.5g8fg0...................\r\nREFERENCE CODE\r\n\r\nA heartbeat is late by the following number of seconds\r\n         120\r\n\r\n以前没遇到过，谁知道是什么原因导致的系统高I/O和内存不足？谢谢\r\n（应用问题可以排除）

huxh

关注中，好象与HACMP有关\r\n我这边也常出，不过好象从来不影响应用，N久以前好象咨询过IBM，给了一个解决方法（更改默认的高/低水平位？这个不能很确定），有没有效果也不记得了

congrd

是HA的问题吗？\r\n这台机器是做了HA,不过与它做双机的另一台机器没有报错呀？！

老农

Description \r\nDeadMan Switch (DMS) close to trigger \r\n\r\nProbable Causes \r\nTopology Services daemon cannot get timely access to CPU \r\n\r\nUser Causes \r\nExcessive I/O load is causing high I/O interrupt traffic \r\nExcessive memory consumption is causing high memory contention \r\n\r\n       Recommended Actions \r\n       Reduce application load on the system \r\n       Change (relax) Topology Services tunable parameters \r\n       Call IBM Service if problem persists \r\n\r\nFailure Causes \r\nProblem in Operating System prevents processes from running \r\nExcessive I/O interrupt traffic prevents processes from running \r\nExcessive virtual memory activity prevents Topology Services from making progress \r\nThe system clock was set forward \r\n\r\n       Recommended Actions \r\n       Examine I/O and memory activity on the system \r\n       Reduce load on the system \r\n       Change (relax) Topology Services tunable parameters \r\n       Call IBM Service if problem persists

congrd

农哥也到这来了。今早LU上不去了，总说我密码不对！\r\n只好发到这来了\r\n你贴的我也看到了，可看不明白\r\n谁能帮我解释一下？谢谢

老农

LU在换论坛程序，要稍等。\r\n\r\n这些信息要求监测你的系统负载情况，还有有些参数是否正确调整过。还有HACMP的补丁版本等。

congrd

看来以后真得多看看了，这台是新机器，P520+Aix5.3+ML02+HACMP5.1+Oracle9I。刚用了2个月，一直都没什么问题，昨天还是第一次报错！

lstars0507

我把这个故障顶起来！\r\n我也是这样的！\r\n这么解决呀，各位仁兄！

o0呼噜Zz

哎哎哎，到处都是农哥，到处都是高手，为啥到处都没有一个解决的办法。。。难道是我还未找到。。。