rabbit-proxy-users Mailing List for RabbIT proxy (Page 39)
Brought to you by:
ernimril
Menu
▾
▴
rabbit-proxy-users — Normal usage and help
You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
(2) |
May
|
Jun
|
Jul
(3) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(4) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 |
Jan
(16) |
Feb
(9) |
Mar
|
Apr
(4) |
May
(7) |
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
(4) |
Dec
(3) |
| 2004 |
Jan
(13) |
Feb
|
Mar
(29) |
Apr
(44) |
May
(17) |
Jun
(14) |
Jul
(7) |
Aug
(2) |
Sep
|
Oct
(2) |
Nov
(2) |
Dec
(26) |
| 2005 |
Jan
(7) |
Feb
(5) |
Mar
|
Apr
(4) |
May
(14) |
Jun
(6) |
Jul
(2) |
Aug
(3) |
Sep
|
Oct
|
Nov
(1) |
Dec
(2) |
| 2006 |
Jan
(14) |
Feb
(6) |
Mar
(11) |
Apr
(7) |
May
(26) |
Jun
(10) |
Jul
(10) |
Aug
(9) |
Sep
(8) |
Oct
(15) |
Nov
(22) |
Dec
(12) |
| 2007 |
Jan
(3) |
Feb
(17) |
Mar
(19) |
Apr
(18) |
May
(13) |
Jun
(11) |
Jul
(16) |
Aug
(14) |
Sep
(1) |
Oct
(5) |
Nov
(38) |
Dec
(4) |
| 2008 |
Jan
(3) |
Feb
(5) |
Mar
(7) |
Apr
(189) |
May
(131) |
Jun
(117) |
Jul
(88) |
Aug
(67) |
Sep
(74) |
Oct
(14) |
Nov
(19) |
Dec
(69) |
| 2009 |
Jan
(32) |
Feb
(23) |
Mar
(35) |
Apr
(47) |
May
(126) |
Jun
(94) |
Jul
(78) |
Aug
(27) |
Sep
(20) |
Oct
(24) |
Nov
(8) |
Dec
(9) |
| 2010 |
Jan
(8) |
Feb
(3) |
Mar
(32) |
Apr
(50) |
May
(88) |
Jun
(59) |
Jul
(36) |
Aug
(43) |
Sep
(25) |
Oct
(2) |
Nov
|
Dec
(2) |
| 2011 |
Jan
|
Feb
(1) |
Mar
(2) |
Apr
(2) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2012 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2013 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(1) |
Oct
|
Nov
|
Dec
|
| 2014 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
|
| 2016 |
Jan
|
Feb
|
Mar
(2) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Costa T. <kt...@ho...> - 2005-01-11 23:02:21
|
> Get the 2.0.36 final release from the normal web pages > http://www.khelekore.org/rabbit/ Thanks! In the mean time I managed to find 2.0.36-pre3 which I installed. Bellow are my results. > Ok, could still be the cache. > Since you do not want any cache in rabbit you could try is to make sure > that > the cache index is not flushed to disk. You can do this by commenting > out the > entire writeCacheIndex in the file NCache.java. The file cache.index is only 159 bytes long and the last modification date is changed per hour, and the whole cache directory size seems too small (a few MB). I guess the above mean that the freezes do not come from the cache. Anyway, if you really want me to test it, I'll be glad to do it. > Ok, that looks like a memory leak is going on. I will have to check > more. > > One thing that you can also do if you still have the console where you > started > java open is to press Control-Backslask (C-\). That will give you a > full > thread > dump of the java vm. That could help, but it will probably be a long > listing, > mail it to me instead of the list. This can occacinally be highly > useful... Ok. I'll do that. In the mean time I had to solve my problem. Since I have a lot of ram I created about 40 rabbit instances and configured the front end proxy servers to load balance them. Normally, one RabbIT was running for just 1-2 minutes before start freezing. The 40 load balanced Rabbits run for several hours now without any problems. Here are a few usefull findings: The table bellow shows a summary of the memory and disk usage of each RabbIT instance. Current Memory and Memory Pool are the second 2 numbers reported by -verbose:gc. RabbIT Port Current Memory Memory Pool Cache Dir 3125 9093K 20576K 1.9M 3126 7099K 41204K 2.0M 3127 10048K 20812K 1.9M 3128 8660K 20920K 2.1M 3129 12199K 23776K 1.9M 3130 6299K 7996K 2.0M 3131 7917K 22624K 2.0M 3132 5367K 8444K 2.1M 3133 5528K 8272K 2.1M 3134 9090K 10036K 2.3M 3135 40180K 42704K 1.9M 3136 7346K 40932K 2.0M 3137 6881K 22368K 2.0M 3138 6373K 7856K 1.9M 3139 8185K 22228K 1.8M 3140 7160K 8008K 1.9M 3141 6380K 8444K 2.2M 3142 8018K 22296K 1.9M 3143 12593K 23756K 2.1M 3144 6741K 7788K 2.0M 3145 7549K 20948K 1.8M 3146 20309K 22252K 1.9M 3147 4718K 8128K 2.0M 3148 7164K 7912K 2.3M 3149 49703K 55352K 1.9M 3150 6505K 8224K 2.2M 3151 34606K 37872K 1.9M 3152 5790K 8096K 2.1M 3153 4734K 8332K 2.0M 3154 20218K 22364K 2.2M 3155 6253K 8148K 2.0M 3156 36062K 39136K 2.3M 3157 48624K 54044K 2.1M 3158 10363K 24048K 2.2M 3159 7874K 23680K 2.0M 3160 7667K 23444K 2.0M 3161 7152K 7984K 1.7M 3162 10297K 25472K 2.0M 3163 33197K 35864K 1.7M 3164 4878K 8312K 2.0M 3165 6265K 8292K 1.8M Each instance runs with: /opt/blackdown-jdk-1.4.2.01/bin/java -Xms2m -Xmx128m -Xss256k -verbose:gc Observations: ------------- 1. Each Rabbit instance has served many more requests in the last hours, than the single Rabbit that was freezing after just 1-2 minutes. This might mean that the "memory leak" is a result of the number of requests per second, or the time needed to release some resource. 2. Although the front end proxies really load balance (round-robin, one by one) all the rabbits, only a few appear to use more ram. 3. Rabbit seems to recover from the "memory leak" in several cases. Check for example instance 3126 and 3136. 4. The memory usage seems to increase in relatively huge steps for the rabbits that use more memory. For example, here is a contigious part of the GC output of instance 3149: [GC 2495K->2014K(2808K), 0.0014060 secs] [GC 2526K->2023K(2808K), 0.0010150 secs] [GC 2535K->2021K(2808K), 0.0009480 secs] [GC 2533K->2025K(2808K), 0.0007940 secs] [GC 2530K->2053K(2808K), 0.0006890 secs] [GC 2565K->2058K(2808K), 0.0009310 secs] [GC 2570K->2091K(2808K), 0.0010140 secs] [GC 2603K->2112K(2808K), 0.0008150 secs] [GC 2573K->2123K(2808K), 0.0009370 secs] [Full GC[Unloading class sun.reflect.GeneratedConstructorAccessor21] [Unloading class sun.reflect.GeneratedMethodAccessor11] [Unloading class sun.reflect.GeneratedConstructorAccessor24] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor16] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor17] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor15] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor13] [Unloading class sun.reflect.GeneratedMethodAccessor10] [Unloading class sun.reflect.GeneratedConstructorAccessor26] [Unloading class sun.reflect.GeneratedConstructorAccessor25] [Unloading class sun.reflect.GeneratedConstructorAccessor27] 2123K->1398K(2808K), 0.0285820 secs] [GC 20534K->20105K(21432K), 0.0013380 secs] [GC 20617K->20147K(21432K), 0.0012960 secs] [GC 20481K->20126K(21432K), 0.0009100 secs] [Full GC[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor18] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor19] 20126K->19837K(21432K), 0.0577450 secs] [GC 41867K->39770K(55352K), 0.0030100 secs] [GC 42074K->39809K(55352K), 0.0050160 secs] [GC 42113K->39878K(55352K), 0.0025770 secs] [GC 42175K->39899K(55352K), 0.0026060 secs] [GC 42203K->39950K(55352K), 0.0032300 secs] [GC 42254K->40025K(55352K), 0.0032420 secs] As shown, the GC logs that "suddenly" the memory usage went from 2MB to 20MB and again suddenly from 20MB to 40MB and stayed there ! On the other hand, instance 3126 which seems that managed to recover, logs this: [GC 2170K->1691K(2832K), 0.0010940 secs] [GC 2203K->1703K(2832K), 0.0010490 secs] [GC 2215K->1725K(2832K), 0.0009540 secs] [GC 2237K->1748K(2832K), 0.0009260 secs] [GC 1928K->1723K(2832K), 0.0008290 secs] [Full GC[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor16] [Unloading class sun.reflect.GeneratedConstructorAccessor23] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor15] [Unloading class sun.reflect.GeneratedConstructorAccessor24] [Unloading class sun.reflect.GeneratedMethodAccessor11] [Unloading class sun.reflect.GeneratedConstructorAccessor25] 1723K->1321K(2832K), 0.0271570 secs] [GC 20083K->19653K(21092K), 0.0014260 secs] [GC 20165K->19696K(21092K), 0.0013250 secs] [GC 20208K->19708K(21092K), 0.0011420 secs] ... (hundreds of logs that gradually increase memory) [GC 41624K->40189K(41908K), 0.0036930 secs] [GC 41725K->40222K(41908K), 0.0028510 secs] [GC 41758K->40303K(41908K), 0.0027120 secs] [GC 41837K->40367K(41908K), 0.0019370 secs] [Full GC[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor22] [Unloading class sun.reflect.GeneratedConstructorAccessor31] [Unloading class sun.reflect.GeneratedMethodAccessor14] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor23] [Unloading class sun.reflect.GeneratedConstructorAccessor32] [Unloading class sun.reflect.GeneratedConstructorAccessor33] [Unloading class sun.reflect.GeneratedMethodAccessor15] [Unloading class sun.reflect.GeneratedConstructorAccessor35] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor21] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor24] [Unloading class sun.reflect.GeneratedConstructorAccessor34] 40367K->5730K(41908K), 0.0836510 secs] [GC 8480K->5967K(41204K), 0.0083550 secs] [GC 8719K->6069K(41204K), 0.0040610 secs] [GC 8821K->6118K(41204K), 0.0054360 secs] [GC 8870K->6105K(41204K), 0.0052020 secs] Here, "Full GC" managed to release the memory back to where it should. Of course after a few hours of work. I guess the above clearly show that there is some kind of direct or indirect memory leak there, and I hope they will help you debug it. Thanks! Costa |
|
From: Robert O. <ro...@kh...> - 2005-01-11 21:21:37
|
Costa Tsaousis wrote: >Hm! I don't know. I'll try 2.0.36-pre3 (where can I get it from?) > > Get the 2.0.36 final release from the normal web pages http://www.khelekore.org/rabbit/ >>Could be the cache cleaner? >>What value do you have for cleanloop? >> >> >I have tried setting size to 0 and loop to 0, to 3600, 86400, etc. >Nothing helped. Currently, I have size and time set to 0 and loop set to >3600. > > Ok, could still be the cache. Since you do not want any cache in rabbit you could try is to make sure that the cache index is not flushed to disk. You can do this by commenting out the entire writeCacheIndex in the file NCache.java. >>Can you check the cache statistics and see if rabbit builds a cache? >>(even if you set it to 0 it will use a small cache for image >>recompression). >> >> >Actually, I have disabled the front ends. Is it possible to get this info >by other means? > > Check the cache dir for files. You can also try to run the CacheStatus program: "java rabbit.cache.CacheStatus /place/to/cache/dir/ " but that program will list all entries in the cache (=> lots of http headers). I think I will make a summar mode in there and let that be the default. >Well, check the log bellow. At the point the memory usage went above >100MB, RabbIT became completely unusable. 100% CPU utilization and just >very few requests served. I also noticed that when RabbIT freezes, the >garbage collector does not run (normally it logs one row every 1-2 >seconds), but when it is frozen, the GC stops logging for several >seconds. > > Ok, that looks like a memory leak is going on. I will have to check more. One thing that you can also do if you still have the console where you started java open is to press Control-Backslask (C-\). That will give you a full thread dump of the java vm. That could help, but it will probably be a long listing, mail it to me instead of the list. This can occacinally be highly useful... /robo |
|
From: Costa T. <kt...@ho...> - 2005-01-11 10:04:56
|
> Do anyone of the proxies in front of rabbit handle the broken chunking > from apache? > If not you should probably try the 2.0.36-pre3 release (will be full > 2.0.36 in short time > if nothing more shows up in my testing). Hm! I don't know. I'll try 2.0.36-pre3 (where can I get it from?) > Could be the cache cleaner? > What value do you have for cleanloop? I have tried setting size to 0 and loop to 0, to 3600, 86400, etc. Nothing helped. Currently, I have size and time set to 0 and loop set to 3600. > How big cache do you have? (I guess 0, since you have other proxies). Right. I don't need a cache for RabbIT. Actually, I restart rabbit once per day. At that time I completely clean any cache it has. > Can you check the cache statistics and see if rabbit builds a cache? > (even if you set it to 0 it will use a small cache for image > recompression). Actually, I have disabled the front ends. Is it possible to get this info by other means? > Could also be GC. Can you try to to add " -verbose:gc " to your > argument > list > and see if it is GC-related?. Well, check the log bellow. At the point the memory usage went above 100MB, RabbIT became completely unusable. 100% CPU utilization and just very few requests served. I also noticed that when RabbIT freezes, the garbage collector does not run (normally it logs one row every 1-2 seconds), but when it is frozen, the GC stops logging for several seconds. This log has been created with -Xmx1536m (64m was giving OutOfMemory errors). [GC 512K->234K(1984K), 0.0039340 secs] [GC 745K->291K(1984K), 0.0023450 secs] [GC 803K->328K(1984K), 0.0034220 secs] [GC 840K->419K(1984K), 0.0036420 secs] [GC 931K->497K(1984K), 0.0028920 secs] [GC 1009K->547K(1984K), 0.0020810 secs] [GC 1059K->613K(1984K), 0.0014280 secs] [GC 1125K->655K(1984K), 0.0011000 secs] [GC 1167K->681K(1984K), 0.0015260 secs] [GC 1193K->736K(1984K), 0.0016470 secs] [GC 1241K->763K(1984K), 0.0011300 secs] [GC 1275K->836K(1984K), 0.0011320 secs] [GC 1348K->949K(1984K), 0.0019430 secs] [GC 1461K->1039K(1984K), 0.0013120 secs] [GC 1550K->1125K(1984K), 0.0018400 secs] [GC 1637K->1156K(1984K), 0.0010890 secs] [GC 1668K->1208K(1984K), 0.0013630 secs] [GC 1720K->1289K(1984K), 0.0013090 secs] [GC 1794K->1347K(1984K), 0.0012780 secs] [GC 1859K->1345K(1984K), 0.0008460 secs] [GC 1857K->1365K(1984K), 0.0009890 secs] [GC 1548K->1344K(1984K), 0.0007000 secs] [Full GC[Unloading class sun.reflect.GeneratedConstructorAccessor3] [Unloading class sun.reflect.GeneratedConstructorAccessor1] [Unloading class sun.reflect.GeneratedConstructorAccessor2] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor1] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor3] [Unloading class sun.reflect.GeneratedMethodAccessor5] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor4] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor2] 1344K->800K(1984K), 0.0196280 secs] [GC 7809K->7374K(8484K), 0.0017450 secs] [GC 7886K->7416K(8484K), 0.0011560 secs] [GC 7928K->7479K(8484K), 0.0016540 secs] [GC 7991K->7582K(8484K), 0.0011850 secs] [GC 8094K->7615K(8484K), 0.0010430 secs] [GC 8127K->7676K(8484K), 0.0013240 secs] [GC 8188K->7730K(8484K), 0.0010260 secs] [GC 8242K->7811K(8484K), 0.0015890 secs] [GC 8323K->7880K(8484K), 0.0015550 secs] [GC 8391K->7916K(8484K), 0.0012330 secs] [GC 8428K->8002K(8612K), 0.0016510 secs] [Full GC[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor5] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor6] 8002K->7775K(8612K), 0.0199320 secs] [GC 8799K->7900K(14112K), 0.0019680 secs] [GC 8924K->7945K(14112K), 0.0016690 secs] [GC 8969K->7982K(14112K), 0.0017250 secs] [GC 9005K->8115K(14112K), 0.0027720 secs] [GC 9131K->8205K(14112K), 0.0018150 secs] [GC 9229K->8253K(14112K), 0.0017360 secs] [GC 9277K->8386K(14112K), 0.0025100 secs] [GC 9410K->8479K(14112K), 0.0022580 secs] [GC 9503K->8527K(14112K), 0.0017070 secs] [GC 9550K->8667K(14112K), 0.0024520 secs] [GC 9691K->8717K(14112K), 0.0017000 secs] [GC 9740K->8776K(14112K), 0.0018140 secs] [GC 9800K->8837K(14112K), 0.0019040 secs] [GC 9859K->8808K(14112K), 0.0016000 secs] [GC 9832K->8874K(14112K), 0.0016230 secs] [GC 9898K->8938K(14112K), 0.0022560 secs] [GC 9962K->8995K(14112K), 0.0019040 secs] [GC 10019K->9001K(14112K), 0.0021020 secs] [GC 10024K->9004K(14112K), 0.0019970 secs] [GC 10028K->9058K(14112K), 0.0016320 secs] [GC 10073K->9094K(14112K), 0.0015690 secs] [GC 10118K->9233K(14112K), 0.0020420 secs] [GC 10257K->9297K(14112K), 0.0025300 secs] [GC 10321K->9367K(14112K), 0.0019540 secs] [GC 10391K->9431K(14112K), 0.0025270 secs] [GC 10455K->9555K(14112K), 0.0027590 secs] [GC 10579K->9611K(14112K), 0.0018400 secs] [GC 10635K->9684K(14112K), 0.0023350 secs] [GC 10707K->9723K(14112K), 0.0021800 secs] [GC 10747K->9770K(14112K), 0.0018520 secs] [GC 10794K->9801K(14112K), 0.0017750 secs] [GC 10825K->9835K(14112K), 0.0017920 secs] [GC 10859K->9823K(14112K), 0.0016860 secs] [GC 10846K->9827K(14112K), 0.0013060 secs] [GC 10851K->9829K(14112K), 0.0011300 secs] [GC 10853K->9837K(14112K), 0.0011300 secs] [GC 10861K->9837K(14112K), 0.0012570 secs] [GC 10861K->9859K(14112K), 0.0015970 secs] [GC 10883K->9870K(14112K), 0.0017110 secs] [GC 10893K->9877K(14112K), 0.0019500 secs] [GC 10900K->9945K(14112K), 0.0021570 secs] [GC 10967K->9967K(14112K), 0.0019390 secs] [GC 10991K->10002K(14112K), 0.0022250 secs] [GC 11023K->10027K(14112K), 0.0018740 secs] [GC 11051K->10135K(14112K), 0.0017790 secs] [GC 11159K->10182K(14112K), 0.0016170 secs] [GC 11206K->10259K(14112K), 0.0018220 secs] [GC 11283K->10254K(14112K), 0.0018570 secs] [GC 11273K->10309K(14112K), 0.0017820 secs] [GC 11332K->10395K(14112K), 0.0017540 secs] [GC 11419K->10472K(14112K), 0.0021140 secs] [GC 11495K->10524K(14112K), 0.0020100 secs] [GC 11548K->10611K(14112K), 0.0017850 secs] [GC 11635K->10620K(14112K), 0.0018190 secs] [GC 11640K->10658K(14112K), 0.0019240 secs] [GC 11681K->10752K(14112K), 0.0017810 secs] [GC 11776K->10828K(14112K), 0.0020860 secs] [GC 11852K->10892K(14112K), 0.0018880 secs] [GC 11906K->10973K(14112K), 0.0019260 secs] [GC 11997K->10993K(14112K), 0.0019050 secs] [GC 12017K->11079K(14112K), 0.0020250 secs] [GC 12096K->11168K(14112K), 0.0031120 secs] [GC 12191K->11247K(14112K), 0.0021610 secs] [GC 12271K->11273K(14112K), 0.0027600 secs] [GC 12297K->11351K(14112K), 0.0033470 secs] [GC 12375K->11371K(14112K), 0.0019710 secs] [GC 12391K->11456K(14112K), 0.0023760 secs] [GC 12479K->11482K(14112K), 0.0020480 secs] [GC 12505K->11592K(14112K), 0.0020170 secs] [GC 12616K->11665K(14112K), 0.0021420 secs] [GC 12688K->11693K(14112K), 0.0029010 secs] [GC 12716K->11784K(14112K), 0.0019910 secs] [GC 12808K->12017K(14112K), 0.0022830 secs] [GC 13041K->12072K(14112K), 0.0023400 secs] [GC 13096K->12168K(14112K), 0.0021880 secs] [GC 13186K->12204K(14112K), 0.0031370 secs] [GC 13228K->12206K(14112K), 0.0018540 secs] [GC 13229K->12249K(14112K), 0.0024660 secs] [GC 13265K->12252K(14112K), 0.0016400 secs] [GC 13275K->12290K(14112K), 0.0018060 secs] [GC 13314K->12332K(14112K), 0.0019640 secs] [GC 13356K->12405K(14112K), 0.0027590 secs] [GC 13428K->12450K(14112K), 0.0019810 secs] [GC 13474K->12533K(14112K), 0.0022940 secs] [GC 13553K->12589K(14112K), 0.0017760 secs] [GC 13612K->12646K(14112K), 0.0026020 secs] [GC 13665K->12677K(14112K), 0.0019050 secs] [GC 13701K->12743K(14112K), 0.0023470 secs] [GC 13767K->12795K(14112K), 0.0023770 secs] [GC 13819K->12882K(14112K), 0.0024370 secs] [GC 13905K->12942K(14112K), 0.0020070 secs] [GC 13965K->12979K(14112K), 0.0021600 secs] [GC 14003K->12968K(14112K), 0.0018170 secs] [GC 13990K->13027K(14112K), 0.0018560 secs] [GC 14051K->13046K(14112K), 0.0022420 secs] [GC 14070K->13134K(14240K), 0.0021590 secs] [Full GC[Unloading class sun.reflect.GeneratedConstructorAccessor5] [Unloading class sun.reflect.GeneratedMethodAccessor7] [Unloading class sun.reflect.GeneratedConstructorAccessor8] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor8] [Unloading class sun.reflect.GeneratedConstructorAccessor6] [Unloading class sun.reflect.GeneratedConstructorAccessor7] [Unloading class sun.reflect.GeneratedConstructorAccessor4] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor9] [Unloading class sun.reflect.GeneratedMethodAccessor6] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor7] [Unloading class sun.reflect.GeneratedSerializationConstructorAccessor10] 13134K->9755K(14240K), 0.0348110 secs] [GC 11226K->9956K(17860K), 0.0035540 secs] [GC 11428K->10074K(17860K), 0.0026080 secs] [GC 11546K->10180K(17860K), 0.0022610 secs] [GC 11645K->10312K(17860K), 0.0023860 secs] [GC 11784K->10505K(17860K), 0.0027650 secs] [GC 11976K->10691K(17860K), 0.0026710 secs] [GC 12163K->10777K(17860K), 0.0021580 secs] [GC 12249K->10919K(17860K), 0.0021360 secs] [GC 12391K->11043K(17860K), 0.0024700 secs] [GC 12515K->11215K(17860K), 0.0026310 secs] [GC 12687K->11399K(17860K), 0.0023470 secs] [GC 12870K->11543K(17860K), 0.0027870 secs] [GC 13005K->11620K(17860K), 0.0021770 secs] [GC 13084K->11726K(17860K), 0.0025340 secs] [GC 13198K->11869K(17860K), 0.0024920 secs] [GC 13341K->12018K(17860K), 0.0024830 secs] [GC 13490K->12095K(17860K), 0.0024170 secs] [GC 13557K->12244K(17860K), 0.0025470 secs] [GC 13716K->12366K(17860K), 0.0026920 secs] [GC 13838K->12588K(17860K), 0.0028810 secs] [GC 14060K->12648K(17860K), 0.0025780 secs] [GC 14120K->12704K(17860K), 0.0023200 secs] [GC 14167K->12892K(17860K), 0.0030100 secs] [GC 14364K->12994K(17860K), 0.0024820 secs] [GC 14466K->13059K(17860K), 0.0041980 secs] [GC 14531K->13175K(17860K), 0.0027810 secs] [GC 14642K->13243K(17860K), 0.0022220 secs] [GC 14715K->13257K(17860K), 0.0023560 secs] [GC 14722K->13372K(17860K), 0.0041560 secs] [GC 14843K->13478K(17860K), 0.0045770 secs] [GC 14950K->13575K(17860K), 0.0027440 secs] [GC 15047K->13692K(17860K), 0.0029840 secs] [GC 15164K->13822K(17860K), 0.0025630 secs] [GC 15294K->14028K(17860K), 0.0027180 secs] [GC 15498K->14165K(17860K), 0.0033270 secs] [GC 15637K->14269K(17860K), 0.0023710 secs] [GC 15741K->14373K(17860K), 0.0040380 secs] [GC 15845K->14331K(17860K), 0.0019270 secs] [GC 15803K->14395K(17860K), 0.0024440 secs] [GC 15867K->14518K(17860K), 0.0027550 secs] [GC 15986K->14683K(17860K), 0.0029100 secs] [GC 16154K->14855K(17860K), 0.0029000 secs] [GC 16318K->14887K(17860K), 0.0027730 secs] [GC 16357K->14971K(17860K), 0.0024210 secs] [GC 16443K->15102K(17860K), 0.0028780 secs] [GC 16574K->15200K(17860K), 0.0024850 secs] [GC 16670K->15311K(17860K), 0.0026940 secs] [GC 16783K->15414K(17860K), 0.0028400 secs] [GC 16886K->15530K(17860K), 0.0045520 secs] [GC 17000K->15665K(17860K), 0.0032010 secs] [GC 17137K->15789K(17860K), 0.0032460 secs] [GC 17261K->15851K(17860K), 0.0028360 secs] [GC 17323K->15931K(17860K), 0.0026640 secs] [GC 17403K->16051K(17860K), 0.0032500 secs] [GC 17523K->16154K(17860K), 0.0048960 secs] [GC 17626K->16255K(17860K), 0.0028620 secs] [GC 17728K->16319K(17860K), 0.0026340 secs] [GC 17791K->16440K(17988K), 0.0025530 secs] [Full GC 16440K->10447K(17988K), 0.0597210 secs] [GC 12047K->10683K(19204K), 0.0040940 secs] [GC 12280K->10734K(19204K), 0.0030960 secs] [GC 12330K->10791K(19204K), 0.0029980 secs] [GC 12391K->10888K(19204K), 0.0028260 secs] [GC 12488K->11025K(19204K), 0.0032710 secs] [GC 12625K->11064K(19204K), 0.0029460 secs] [GC 12662K->11092K(19204K), 0.0026510 secs] [GC 12692K->11186K(19204K), 0.0034690 secs] [GC 12786K->11261K(19204K), 0.0033740 secs] [GC 12861K->11324K(19204K), 0.0036640 secs] [GC 12924K->11329K(19204K), 0.0027210 secs] [GC 12928K->11386K(19204K), 0.0031880 secs] [GC 12985K->11464K(19204K), 0.0030650 secs] [GC 13062K->11512K(19204K), 0.0029720 secs] [GC 13112K->11545K(19204K), 0.0029240 secs] [GC 13145K->11597K(19204K), 0.0033030 secs] [GC 13197K->11661K(19204K), 0.0029980 secs] [GC 13261K->11669K(19204K), 0.0030330 secs] [GC 13268K->11757K(19204K), 0.0033080 secs] [GC 13355K->11782K(19204K), 0.0028600 secs] [GC 13382K->11863K(19204K), 0.0032340 secs] [GC 13463K->11952K(19204K), 0.0033340 secs] [GC 13551K->12016K(19204K), 0.0031490 secs] [GC 13609K->12088K(19204K), 0.0032420 secs] [GC 13685K->12177K(19204K), 0.0032320 secs] [GC 13777K->12241K(19204K), 0.0031310 secs] [GC 13841K->12296K(19204K), 0.0031240 secs] [GC 13895K->12367K(19204K), 0.0033750 secs] [GC 13962K->12463K(19204K), 0.0034730 secs] [GC 14063K->12561K(19204K), 0.0034150 secs] [GC 14161K->12652K(19204K), 0.0033480 secs] [GC 14252K->12675K(19204K), 0.0033670 secs] [GC 14275K->12811K(19204K), 0.0037510 secs] [GC 14411K->12947K(19204K), 0.0036320 secs] [GC 14547K->12974K(19204K), 0.0032380 secs] [GC 14574K->13025K(19204K), 0.0035040 secs] [GC 14625K->13123K(19204K), 0.0034620 secs] [GC 14723K->13217K(19204K), 0.0035630 secs] [GC 14817K->13258K(19204K), 0.0033890 secs] [GC 14859K->13294K(19204K), 0.0056130 secs] [GC 14894K->13350K(19204K), 0.0032090 secs] [GC 14948K->13489K(19204K), 0.0035530 secs] [GC 15089K->13549K(19204K), 0.0035310 secs] [GC 15148K->13599K(19204K), 0.0032650 secs] [GC 15199K->13700K(19204K), 0.0033100 secs] [GC 15298K->13802K(19204K), 0.0036500 secs] [GC 15402K->13950K(19204K), 0.0032740 secs] [GC 15560K->14015K(19204K), 0.0034990 secs] [GC 15615K->13979K(19204K), 0.0031580 secs] [GC 15576K->14052K(19204K), 0.0035900 secs] [GC 15652K->14034K(19204K), 0.0034410 secs] [GC 15634K->14051K(19204K), 0.0030440 secs] [GC 15644K->14129K(19204K), 0.0034930 secs] [GC 15729K->14169K(19204K), 0.0036320 secs] [GC 15769K->14220K(19204K), 0.0029380 secs] [GC 15817K->14305K(19204K), 0.0036340 secs] [GC 15900K->14433K(19204K), 0.0038610 secs] [GC 16032K->14524K(19204K), 0.0036200 secs] [GC 16124K->14601K(19204K), 0.0035810 secs] [GC 16199K->14652K(19204K), 0.0037040 secs] [GC 16252K->14754K(19204K), 0.0036340 secs] [GC 16354K->14884K(19204K), 0.0038260 secs] [GC 16484K->14908K(19204K), 0.0037320 secs] [GC 16499K->14945K(19204K), 0.0036060 secs] [GC 16537K->15016K(19204K), 0.0036160 secs] [GC 16616K->15106K(19204K), 0.0042390 secs] [GC 16706K->15237K(19204K), 0.0041370 secs] [GC 16836K->15325K(19204K), 0.0034910 secs] [GC 16927K->15403K(19204K), 0.0036210 secs] [GC 17003K->15428K(19204K), 0.0081670 secs] [GC 17028K->15467K(19204K), 0.0036350 secs] [GC 17067K->15507K(19204K), 0.0036630 secs] [GC 17107K->15513K(19204K), 0.0035120 secs] [GC 17113K->15485K(19204K), 0.0036300 secs] [GC 17085K->15491K(19204K), 0.0034370 secs] [GC 17083K->15526K(19204K), 0.0034540 secs] [GC 17125K->15557K(19204K), 0.0050430 secs] [GC 17153K->15589K(19204K), 0.0035980 secs] [GC 17187K->15601K(19204K), 0.0034010 secs] [GC 17201K->15622K(19204K), 0.0035320 secs] [GC 17222K->15700K(19204K), 0.0037730 secs] [GC 17291K->15737K(19204K), 0.0040130 secs] [GC 17337K->15751K(19204K), 0.0066420 secs] [GC 17348K->15749K(19204K), 0.0034200 secs] [GC 17339K->15743K(19204K), 0.0034060 secs] [GC 17343K->15806K(19204K), 0.0035900 secs] [GC 17400K->15810K(19204K), 0.0034230 secs] [GC 17410K->15788K(19204K), 0.0033930 secs] [GC 17388K->15828K(19204K), 0.0036850 secs] [GC 17427K->15858K(19204K), 0.0037180 secs] [GC 17452K->15883K(19204K), 0.0039180 secs] [GC 17483K->15931K(19204K), 0.0035940 secs] [GC 17531K->15981K(19204K), 0.0035560 secs] [GC 17581K->16027K(19204K), 0.0036940 secs] [GC 17618K->16151K(19204K), 0.0044930 secs] [GC 17751K->16226K(19204K), 0.0043750 secs] [GC 17825K->16301K(19204K), 0.0038530 secs] [GC 17896K->16340K(19204K), 0.0034640 secs] [GC 17938K->16363K(19204K), 0.0063920 secs] [GC 17960K->16406K(19204K), 0.0037640 secs] [GC 18005K->16446K(19204K), 0.0035670 secs] [GC 18040K->16540K(19204K), 0.0035410 secs] [GC 18140K->16529K(19204K), 0.0035480 secs] [GC 18124K->16575K(19204K), 0.0059970 secs] [GC 18175K->16579K(19204K), 0.0067210 secs] [GC 18179K->16669K(19204K), 0.0048300 secs] [GC 18269K->16737K(19204K), 0.0076220 secs] [GC 18337K->16793K(19204K), 0.0039120 secs] [GC 18393K->16861K(19204K), 0.0072320 secs] [GC 18453K->16916K(19204K), 0.0041300 secs] [GC 18516K->16981K(19204K), 0.0038480 secs] [GC 18581K->17044K(19204K), 0.0039720 secs] [GC 18644K->17147K(19204K), 0.0080300 secs] [GC 18747K->17230K(19204K), 0.0104950 secs] [GC 18827K->17365K(19204K), 0.0093330 secs] [GC 18965K->17576K(19204K), 0.0099360 secs] [GC 19176K->17724K(19332K), 0.2783080 secs] [Full GC 17724K->12334K(19332K), 0.0969750 secs] [GC 14382K->12776K(22864K), 0.0114740 secs] [GC 14824K->12939K(22864K), 0.0150260 secs] [GC 14987K->13125K(22864K), 0.0104470 secs] [GC 15173K->13111K(22864K), 0.0043060 secs] [GC 15159K->13156K(22864K), 0.0040750 secs] [GC 15199K->13253K(22864K), 0.0039230 secs] [GC 15301K->13294K(22864K), 0.0061790 secs] [GC 15339K->13280K(22864K), 0.0039790 secs] [GC 15328K->13362K(22864K), 0.0039190 secs] [GC 15410K->13434K(22864K), 0.0053220 secs] [GC 15480K->13459K(22864K), 0.0060960 secs] [GC 15507K->13524K(22864K), 0.0077810 secs] [GC 15569K->13467K(22864K), 0.0041180 secs] [GC 15515K->13543K(22864K), 0.0040020 secs] [GC 15582K->13621K(22864K), 0.0041150 secs] [GC 15664K->13674K(22864K), 0.0043610 secs] [GC 15722K->13753K(22864K), 0.0044410 secs] [GC 15801K->13766K(22864K), 0.0093460 secs] [GC 15814K->13841K(22864K), 0.0042710 secs] [GC 15889K->13858K(22864K), 0.0040020 secs] [GC 15905K->13916K(22864K), 0.0040660 secs] [GC 15964K->13928K(22864K), 0.0043220 secs] [GC 15976K->13946K(22864K), 0.0041480 secs] [GC 15994K->14103K(22864K), 0.0045940 secs] [GC 16151K->14242K(22864K), 0.0048350 secs] [GC 16287K->14324K(22864K), 0.0044560 secs] [GC 16372K->14487K(22864K), 0.0044490 secs] [GC 16535K->14742K(22864K), 0.0087380 secs] [GC 16790K->14785K(22864K), 0.0047370 secs] [GC 16826K->14842K(22864K), 0.0046690 secs] [GC 16889K->14977K(22864K), 0.0044480 secs] [GC 17025K->15033K(22864K), 0.0041880 secs] [GC 17079K->15095K(22864K), 0.0042140 secs] [GC 17143K->15235K(22864K), 0.0041030 secs] [GC 17283K->15265K(22864K), 0.0047150 secs] [GC 17313K->15401K(22864K), 0.0053060 secs] [GC 17449K->15518K(22864K), 0.0047610 secs] [GC 17566K->15644K(22864K), 0.0053130 secs] [GC 17692K->15696K(22864K), 0.0045630 secs] [GC 17744K->15813K(22864K), 0.0047480 secs] [GC 17859K->15796K(22864K), 0.0042540 secs] [GC 17842K->15927K(22864K), 0.0047670 secs] [GC 17970K->16013K(22864K), 0.0049730 secs] [GC 18061K->16099K(22864K), 0.0050140 secs] [GC 18147K->16324K(22864K), 0.0057410 secs] [GC 18372K->16520K(22864K), 0.0051360 secs] [GC 18568K->16577K(22864K), 0.0045540 secs] [GC 18624K->16754K(22864K), 0.0049260 secs] [GC 18802K->16881K(22864K), 0.0049020 secs] [GC 18929K->17030K(22864K), 0.0048870 secs] [GC 19078K->17163K(22864K), 0.0050820 secs] [GC 19203K->17232K(22864K), 0.0046480 secs] [GC 19279K->17280K(22864K), 0.0043290 secs] [GC 19328K->17358K(22864K), 0.0099010 secs] [GC 19406K->17400K(22864K), 0.0047810 secs] [GC 19448K->17489K(22864K), 0.0045130 secs] [GC 19537K->17619K(22864K), 0.0050730 secs] [GC 19667K->17618K(22864K), 0.0042940 secs] [GC 19666K->17707K(22864K), 0.0043890 secs] [GC 19755K->17750K(22864K), 0.0049030 secs] [GC 19798K->17848K(22864K), 0.0045990 secs] [GC 19896K->17901K(22864K), 0.0044070 secs] [GC 19948K->17939K(22864K), 0.0053280 secs] [GC 19987K->18030K(22864K), 0.0049060 secs] [GC 20074K->18237K(22864K), 0.0051640 secs] [GC 20283K->18330K(22864K), 0.0046130 secs] [GC 20378K->18472K(22864K), 0.0052890 secs] [GC 20520K->18568K(22864K), 0.0045550 secs] [GC 20616K->18570K(22864K), 0.0045910 secs] [GC 20617K->18614K(22864K), 0.0048650 secs] [GC 20662K->18705K(22864K), 0.0044690 secs] [GC 20753K->18834K(22864K), 0.0047370 secs] [GC 20882K->18962K(22864K), 0.0054050 secs] [GC 21010K->19128K(22864K), 0.0046420 secs] [GC 21174K->19285K(22864K), 0.0052510 secs] [GC 21333K->19334K(22864K), 0.0050150 secs] [GC 21382K->19463K(22864K), 0.0051450 secs] [GC 21509K->19478K(22864K), 0.0050760 secs] [GC 21526K->19493K(22864K), 0.0054140 secs] [GC 21541K->19600K(22864K), 0.0047440 secs] [GC 21648K->19700K(22864K), 0.0048700 secs] [GC 21748K->19718K(22864K), 0.0046540 secs] [GC 21766K->19713K(22864K), 0.0046040 secs] [GC 21761K->19829K(22864K), 0.0058690 secs] [GC 21877K->20046K(22864K), 0.0105550 secs] [GC 22093K->20233K(22864K), 0.0064990 secs] [GC 22281K->20323K(22864K), 0.0052270 secs] [GC 21401K->20334K(22864K), 0.0043700 secs] [Full GC 20334K->13844K(22864K), 0.0717630 secs] [GC 22898K->20665K(25828K), 0.0069570 secs] [GC 23156K->20830K(25828K), 0.0055670 secs] [GC 23326K->20926K(25828K), 0.0056010 secs] [GC 23422K->21134K(25828K), 0.0057690 secs] [GC 23351K->21224K(25828K), 0.0057390 secs] [Full GC 21224K->18933K(25828K), 0.0738370 secs] [GC 28608K->25985K(34948K), 0.0062890 secs] [GC 29056K->26082K(34948K), 0.0075950 secs] [GC 29144K->26161K(34948K), 0.0071200 secs] [GC 29230K->26217K(34948K), 0.0062510 secs] [GC 29288K->26318K(34948K), 0.0060720 secs] [GC 29390K->26398K(34948K), 0.0057490 secs] [GC 29470K->26442K(34948K), 0.0100110 secs] [GC 29513K->26521K(34948K), 0.0055260 secs] [GC 28762K->26592K(34948K), 0.0055340 secs] [Full GC 26592K->24584K(34948K), 0.0943330 secs] [GC 46276K->43433K(63564K), 0.0061710 secs] [Full GC 43433K->43429K(63564K), 0.0561070 secs] [GC 67440K->62218K(78848K), 0.0073480 secs] [GC 67978K->62349K(78848K), 0.0083420 secs] [GC 68109K->62489K(78848K), 0.0088360 secs] [GC 68248K->62647K(78848K), 0.0093530 secs] [GC 63122K->62121K(78848K), 0.0044700 secs] [Full GC 62121K->62121K(78848K), 0.0572980 secs] [GC 82917K->80927K(112304K), 0.0057840 secs] [Full GC 80927K->73235K(112304K), 0.1588800 secs] [GC 100581K->92469K(132236K), 0.0101350 secs] [GC 101557K->92542K(132236K), 0.0109990 secs] [GC 101628K->92761K(132236K), 0.0099430 secs] [GC 97487K->92554K(132236K), 0.0133420 secs] [Full GC 92554K->91768K(132236K), 0.2328910 secs] [GC 114541K->110630K(165492K), 0.0073170 secs] [Full GC 110630K->110626K(165492K), 0.0608510 secs] [GC 142133K->130202K(199228K), 0.0146400 secs] [GC 143442K->130343K(199228K), 0.0204090 secs] [GC 133386K->129651K(199228K), 0.0091440 secs] [Full GC 129651K->129651K(199228K), 0.0690250 secs] [GC 150117K->148360K(233368K), 0.0069660 secs] [Full GC 148360K->148360K(233368K), 0.0672830 secs] [GC 184027K->168339K(266852K), 0.0170140 secs] [GC 175384K->167759K(266852K), 0.0110320 secs] [Full GC 167759K->159480K(266852K), 0.3219960 secs] [GC 178232K->178030K(286860K), 0.0355170 secs] [GC 179157K->178114K(286860K), 0.0049690 secs] [GC 196866K->196333K(286860K), 0.0362860 secs] [GC 215079K->198156K(286860K), 0.0168510 secs] [GC 199231K->196734K(286860K), 0.0070810 secs] [GC 214998K->214711K(286860K), 0.0370000 secs] [GC 233463K->233343K(286860K), 0.0369330 secs] [GC 252095K->234899K(286860K), 0.0179200 secs] [GC 247850K->234560K(286860K), 0.0256440 secs] [GC 253312K->251750K(286860K), 0.0372250 secs] [GC 270500K->270345K(305484K), 0.0510770 secs] [Full GC 270345K->270345K(305484K), 0.0942260 secs] [GC 298303K->272901K(485520K), 0.0247080 secs] [GC 292040K->289100K(485520K), 0.0395150 secs] [GC 320204K->308801K(485520K), 0.0541180 secs] [GC 334175K->327025K(485520K), 0.0530490 secs] [GC 358129K->347239K(485520K), 0.0716530 secs] [GC 361156K->347768K(485520K), 0.0292580 secs] [GC 398598K->386819K(485520K), 0.0705530 secs] [GC 417923K->406672K(485520K), 0.0512470 secs] |
|
From: Robert O. <ro...@kh...> - 2005-01-10 23:55:32
|
Costa Tsaousis wrote: >This setup is serving about 2.5 mil requests per day. > > Ok, cool. Do anyone of the proxies in front of rabbit handle the broken chunking from apache? If not you should probably try the 2.0.36-pre3 release (will be full 2.0.36 in short time if nothing more shows up in my testing). >I have the following issue: > >There must be some recurring job inside java that freezes RabbIT every a >few minutes, especially during peak hours. > > Could be the cache cleaner? What value do you have for cleanloop? How big cache do you have? (I guess 0, since you have other proxies). Can you check the cache statistics and see if rabbit builds a cache? (even if you set it to 0 it will use a small cache for image recompression). Could also be GC. Can you try to to add " -verbose:gc " to your argument list and see if it is GC-related?. >Bellow, is the output of vmstat that shows the problem. > Ill look more at that later on. It would also be interesting to know if java/5 is any better. I do not believe that it will make any change, but if you have the option to try it (and I guess that you can have a hard time trying it). /robo |
|
From: Costa T. <kt...@ho...> - 2005-01-10 23:28:27
|
Hi, I am using RabbIT 2.0.35 with blackdown-jdk 1.4.2.01 on Gentoo Linux. RabbIT is installed on 2x 3GHz Xeon with 4GB RAM, and it has other proxies in front (clients' requests) and behind it (server replies). RabbIT is run with this command: /opt/blackdown-jdk-1.4.2.01/bin/java -Xms2m -Xmx64m -Xss256k -classpath /opt/RabbIT2/current rabbit.proxy.Proxy This setup is serving about 2.5 mil requests per day. I have the following issue: There must be some recurring job inside java that freezes RabbIT every a few minutes, especially during peak hours. Bellow, is the output of vmstat that shows the problem. Follow the 'us' (CPU/user) column at the rows where it shows 100. At that point the 'r' (PROCS/ready) column shows that there are dozens of processes ready but there is no CPU available to serve them. The machine is running only RabbIT. I have tried using -Xincgc (incremental garbage collector) but then RabbIT exits with something like: Exception in thread "RabbIT2: 193" java.lang.OutOfMemoryError: requested 24 bytes for CHeapObj-new. Out of swap space? I have tried increasing -Xms (512MB) and -Xmx (1,7GB) but the problem gets worse. The problem appears only during peak times. Any help is appreciated. procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa 2 0 0 1852248 429796 765508 0 0 0 984 1980 656 8 5 79 8 0 0 0 1854488 429796 765644 0 0 0 0 2028 685 8 6 86 0 10 0 0 1838872 429796 765780 0 0 0 0 2527 1466 18 14 68 0 23 0 0 1811480 429796 765916 0 0 0 0 2889 3319 60 40 0 0 1 0 0 1851816 429796 765848 0 0 0 0 2711 2911 24 12 64 0 0 0 0 1853176 429796 765848 0 0 0 960 2174 799 9 5 76 10 0 0 0 1852712 429796 765916 0 0 0 0 2199 969 17 11 72 0 0 0 0 1853088 429796 765984 0 0 0 0 2178 807 8 6 87 0 21 0 0 1844416 429796 766188 0 0 0 0 2510 1394 32 17 52 0 7 0 0 1835736 429796 766188 0 0 0 0 1905 2449 89 11 0 0 2 0 0 1838736 429796 766324 0 0 0 868 2570 3340 46 29 20 4 0 0 0 1851824 429796 766324 0 0 0 0 2377 979 14 10 76 0 1 0 0 1851952 429796 766324 0 0 0 0 2020 747 5 4 91 0 1 0 0 1849176 429796 766324 0 0 0 0 2134 860 14 10 76 0 25 0 0 1842528 429804 766452 0 0 0 0 2464 848 55 6 40 0 1 0 0 1849184 429804 766860 0 0 0 1316 2766 21895 57 34 8 0 28 0 0 1844112 429804 766928 0 0 0 0 2233 980 88 7 6 0 80 0 0 1844224 429804 766928 0 0 0 0 1766 565 100 0 0 0 61 0 0 1844224 429804 766928 0 0 0 0 1510 315 100 0 0 0 61 0 0 1844160 429804 766928 0 0 0 0 1556 235 100 0 0 0 12 0 0 1847888 429804 767132 0 0 0 580 2004 15191 93 6 0 0 12 0 0 1847888 429804 767132 0 0 0 0 1430 167 100 0 0 0 11 0 0 1848536 429804 767132 0 0 0 0 1356 181 100 0 0 0 15 0 0 1844312 429804 767200 0 0 0 0 1441 2540 93 5 2 0 65 1 0 1848568 429804 767200 0 0 0 884 2539 1669 96 4 0 0 63 0 0 1848760 429804 767268 0 0 0 292 1602 294 100 0 0 0 97 0 0 1848824 429804 767268 0 0 0 0 1504 288 100 0 0 0 10 0 0 1838664 429804 766928 0 0 0 0 3150 12033 62 37 1 0 9 0 0 1845464 429804 767064 0 0 0 0 3824 28732 66 31 3 0 56 1 0 1836752 429804 766928 0 0 0 1020 2147 3557 91 8 0 0 14 0 0 1837936 429804 766860 0 0 0 308 1955 4486 91 9 0 0 11 0 0 1821648 429804 766860 0 0 0 0 2411 3643 63 37 0 0 1 0 0 1848680 429804 766792 0 0 0 0 1971 787 18 10 72 0 1 0 0 1848680 429804 766860 0 0 0 0 2245 1164 29 20 51 0 1 1 0 1850536 429804 766928 0 0 0 560 2152 876 8 6 83 3 2 0 0 1849448 429804 766860 0 0 0 296 2249 920 10 7 80 2 1 0 0 1850408 429804 766928 0 0 0 0 2117 901 10 8 81 0 0 0 0 1850728 429804 766928 0 0 0 0 2230 829 9 6 85 0 1 0 0 1848552 429804 766996 0 0 0 0 1914 551 4 3 92 0 1 1 0 1843576 429804 766996 0 0 0 532 1840 680 7 5 86 2 1 0 0 1850424 429808 767060 0 0 0 284 2190 875 7 6 85 2 0 0 0 1850424 429808 767196 0 0 0 0 1881 711 12 8 81 0 9 0 0 1837592 429808 767196 0 0 0 0 1979 944 12 10 78 0 0 0 0 1849976 429808 767332 0 0 0 0 1966 1056 36 22 42 0 0 1 0 1850232 429808 767332 0 0 0 544 2097 833 5 4 89 2 0 0 0 1849976 429808 767400 0 0 0 272 2035 779 9 5 83 3 0 0 0 1850680 429808 767400 0 0 0 0 1915 661 7 5 89 0 -- Costa |
|
From: Robert O. <ro...@kh...> - 2004-12-31 12:18:08
|
George Dimitriadis wrote:
> How can I disable or customize the "504 Gateway Time-out" Page......
> Any help would be appreciated.....
You can not easily disable it.
If you want to modify it you currently have to change the code.
Check StandardResponseHeaders.java the code should be quite
easy to understand.
I plan to make it possible to use files instead of the hard coded
versions, but not sure when I will have time to do that.
It may seem simple, but there has to be support for changing the
data on the page ("got a 500 internal error on $page" in file =>
"got a 500 internal error on http://slashdot.org" or something
similar)
/robo
|
|
From: George D. <gdi...@ho...> - 2004-12-31 11:25:20
|
I've set it to 50GB......(50000 MB) --------------------------------------------------------------------------------------------------- # The time in hours to cache files, unless specified otherwise (in the # http header that is). cachetime=168 # The maximal size of the proxy in MB. # The cache sweeps at regular intervalls and if the cache is bigger # some stuff is cleaned out. maxsize=50000 # The time the cleaner sleeps between cleanups. # time is in seconds. cleanloop=300 --------------------------------------------------------------------------------------------------- >From: Robert Olofsson <ro...@kh...> >To: George Dimitriadis <gdi...@ho...> >CC: rab...@li... >Subject: Re: [Rabbit-proxy-users] Rabbit RHE3, 2jre1.4.2_06, RabbIT2-2.0.35 >errors >Date: Fri, 31 Dec 2004 12:18:57 +0100 > >George Dimitriadis wrote: > >>[20/Dec/2004:16:37:34 GMT][DEBUG][removing cache file due to exception: >>rabbit.io.MaximumSizeExcededException: resource size got too >>big] > >How large cache do you have? If you have set the cache to something like 1 >MB this is likely to happen. > >/robo > > > >------------------------------------------------------- >The SF.Net email is sponsored by: Beat the post-holiday blues >Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. >It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt >_______________________________________________ >Rabbit-proxy-users mailing list >Rab...@li... >https://lists.sourceforge.net/lists/listinfo/rabbit-proxy-users _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ |
|
From: Robert O. <ro...@kh...> - 2004-12-31 11:19:06
|
George Dimitriadis wrote: > [20/Dec/2004:16:37:34 GMT][DEBUG][removing cache file due to > exception: rabbit.io.MaximumSizeExcededException: resource size got too > big] How large cache do you have? If you have set the cache to something like 1 MB this is likely to happen. /robo |
|
From: George D. <gdi...@ho...> - 2004-12-31 11:14:13
|
How can I disable or customize the "504 Gateway Time-out" Page...... Any help would be appreciated..... George D. _________________________________________________________________ FREE pop-up blocking with the new MSN Toolbar - get it now! http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/ |
|
From: George D. <gdi...@ho...> - 2004-12-31 09:12:13
|
I receive the following errors os some sites: [20/Dec/2004:16:37:34 GMT][DEBUG][removing cache file due to exception: rabbit.io.MaximumSizeExcededException: resource size got too big] [20/Dec/2004:16:37:34 GMT][DEBUG][removing cache file due to exception: rabbit.io.MaximumSizeExcededException: resource size got too big] In this case some images are not displayed. When I refresh the prefious not displayes images may or may not be diplayed. My configuration is: Red Hat Enterprise Linux AS release 3 (Taroon Update 3) 2jre1.4.2_06 RabbIT2-2.0.35 Any ideas what is causing this problem ........Any help would be appreciated..... George D. _________________________________________________________________ Don't just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/ |
|
From: Robert O. <ro...@kh...> - 2004-12-30 17:21:09
|
Holger Krull wrote: >> http://www.khelekore.org/rabbit/RabbIT2.0.36-pre2.tar.gz > > Hello, > this version solved my recompression problems. Odd. There should be nothing new in there to handle that. > And i get less unable to parse expire date messages, Well, that could have been fixed by a data parsing bug fix, good to hear that it seems to help. > instead a lot [WARN][Bad conversion: /dev/shm/rcache/temp/794 show up. > But the pictures get recompressed. Hmmm. That happens when the return value of the image conversion is non-zero. On unix that usually that indicates an error, on windows it seems that return values are totaly random. I have added the exit value to that warning. Happy new year! /robo |
|
From: Holger K. <hol...@gm...> - 2004-12-30 15:28:59
|
> Ok, can you try this one: > > http://www.khelekore.org/rabbit/RabbIT2.0.36-pre2.tar.gz > Hello, this version solved my recompression problems. And i get less unable to parse expire date messages, instead a lot [WARN][Bad conversion: /dev/shm/rcache/temp/794 show up. But the pictures get recompressed. Thanks, Holger |
|
From: Robert O. <ro...@kh...> - 2004-12-29 21:35:48
|
Leos Buchacek wrote: > I will appreciate any other hints, Ok, can you try this one: http://www.khelekore.org/rabbit/RabbIT2.0.36-pre2.tar.gz You will have to configure it to use the DNSSunHandler (in conf/rabbit.conf): dnsHandler=rabbit.proxy.DNSSunHandler /robo |
|
From: <Leo...@se...> - 2004-12-29 08:53:30
|
Hallo, thank You very much for your reply. > zsh for windows? > Can you try it in the standard command.com (or is that cmd?). > I am not sure, but I would guess that dnsjava does not understand > how it should do name lookups... Yes, I tried that in standard "cmd" thing in XP, unfortunately with the absolute same result (which I described in my first post). I would like to mention, that I've complete new instalation on my pc, so I think that XP's reinstalation isn't way for solution (but I'll try that as well - probably my java instalation only looks on first spot like well-instaled(?)). I'am afraid, that I can't see any other ways for solution... NOTE: I'am using zsh and gnu-tools on all of my computers (XP, Irix and OpenBSD), because of the same user interface. I will appreciate any other hints, thanks in advance Leos |
|
From: Robert O. <ro...@kh...> - 2004-12-28 18:46:48
|
Leos Buchacek wrote: >zsh for windows (CLI) >Windows XP Prof > > zsh for windows? Can you try it in the standard command.com (or is that cmd?). I am not sure, but I would guess that dnsjava does not understand how it should do name lookups.. /robo |
|
From: <Leo...@se...> - 2004-12-28 09:10:08
|
Hallo, On my standalone pc, I used 'scache' web chache [http://scache.sourceforge.net/] but there are some security problems and now I want to try RabbIT. I'am only user and newbie in Java world, so I can't catch my problem and its solution. RabbIT is alive and running, but every connection is ending with 504 Gateway Time-out Unable to handle request: Unable to connect to URL: http://www.google.com/search?q=bus&sourceid=opera&num=0&ie=utf-8&oe=utf-8 I feel that problem will be in my java instalation on win XP, but every my tests (e.g. exporting CLASSPATH [RabbIT's README hint] in zsh, which I'am using on win XP for these tasks) are without positive result, so far. Any hint for my newbie problem will be appreciated Thanks in advance ---------------------------------------------------------------------- Error to google [snip from error_log_onlycaching]: [28/XII/2004:07:41:05 GMT][WARN][Failed to connect socket:java.io.IOException: Could not connect to host: java.net.UnknownHostException: unknown host] [28/XII/2004:07:41:35 GMT][WARN][Could not find the host: 'java.net.UnknownHostException: unknown host': GET http://www.google.com/search?q=vlak&sourceid=opera&num=0&ie=utf-8&oe=utf-8 HTTP/1.1 User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; cs) Opera 7.60 Host: www.google.com Accept: text/html, application/xml;q=0.9, application/xhtml+xml, image/png, image/jpeg, image/gif, image/x-xbitmap, */*;q=0.1 Accept-Language: cs;q=1.0,en;q=0.9 Accept-Charset: iso-8859-2, utf-8, utf-16, iso-8859-1;q=0.6, *;q=0.1 Accept-Encoding: deflate, gzip, x-gzip, identity, *;q=0 Cookie: PREF=ID=1520b61cc1cd190a:TM=1103296867:LM=1103296867:S=yss00_QK66_Omxeb Cookie2: $Version=1 Via: HTTP/1.0 RabbIT Error to logrotate [snip from error_log_onlycaching]: [28/XII/2004:07:45:35 GMT][WARN][Failed to connect socket:java.io.IOException: Could not connect to host: java.net.UnknownHostException: unknown host] [28/XII/2004:07:45:35 GMT][WARN][strange error setting up web connection: java.io.IOException: Unable to connect to URL: http://www.google.com/favicon.ico] [28/XII/2004:07:51:20 GMT][MSG][Configuration loaded, ready for action.] [28/XII/2004:07:51:20 GMT][MSG][Started] [28/XII/2004:07:51:41 GMT][WARN][Couldnt find class:favicon.ico, java.lang.ClassNotFoundException: favicon.ico] [28/XII/2004:08:00:33 GMT][MSG][log rotation requested.] [28/XII/2004:08:00:33 GMT][ERROR][failed to rotate error log!] [28/XII/2004:08:00:33 GMT][ERROR][failed to rotate access log!] [28/XII/2004:08:07:57 GMT][MSG][Configuration loaded, ready for action.] [28/XII/2004:08:07:57 GMT][MSG][Started] [28/XII/2004:08:08:52 GMT][WARN][Could not find the host: 'java.net.UnknownHostException: unknown host': ET http://www.google.com/search?q=bus&sourceid=opera&num=0&ie=utf-8&oe=utf-8 HTTP/1.1 User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; cs) Opera 7.60 Host: www.google.com Accept: text/html, application/xml;q=0.9, application/xhtml+xml, image/png, image/jpeg, image/gif, image/x-xbitmap, */*;q=0.1 Accept-Language: cs;q=1.0,en;q=0.9 Accept-Charset: iso-8859-2, utf-8, utf-16, iso-8859-1;q=0.6, *;q=0.1 Accept-Encoding: deflate, gzip, x-gzip, identity, *;q=0 Cookie: PREF=ID=1520b61cc1cd190a:TM=1103296867:LM=1103296867:S=yss00_QK66_Omxeb Cookie2: $Version=1 Via: HTTP/1.0 RabbIT My configuration: ---------------------------------------------------------------------- Version: RabbIT proxy version 2.0.35 Running on: w/IP : 9668 Started at: Tue Dec 28 09:07:57 CET 2004 Current time: Tue Dec 28 09:13:10 CET 2004 Alive and kicking with 1 current connections. java -version java version "1.4.2" Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.2-b28) Java HotSpot(TM) Client VM (build 1.4.2-b28, mixed mode) zsh for windows (CLI) Windows XP Prof browser Opera 7.60 for win ---------------------------------------------------------------------- Thanks in advance Leos |
|
From: Holger K. <hol...@gm...> - 2004-12-18 23:19:52
|
I did some more testing, different systems and with IBMs virtual machine Conclusion: this thing is a suse problem. Suse 9.1 and 9.2 have it, redhat and debian don't. (The date parsing problems also occour only on suse) Do you have any idea how the kernel config or the surrounding system could influence the decision to recompress the image? Holger |
|
From: Holger K. <hol...@gm...> - 2004-12-16 04:02:48
|
The admin pages load perfectly with only one time entering of the password if i use noproxy.murpel:8080. > Since the problem seems to be with some gif images I have to ask if you > have libgif and/or libungif installed on both systems? /usr/lib/libungif.so.4.1.3 /usr/lib/libgif.so.4.1.3 on both systems. It's not only gifs. Jpegs on gmx.de have the same problem. Holger |
|
From: Robert O. <ro...@kh...> - 2004-12-15 23:39:53
|
Holger Krull wrote: > 2.0.35 I tried 2.0.26 with the same result. Ok, I really have to recomend that you use 2.0.35. > The server in question has serveral ip adresses and dns names. Maybe > that's confusing. That is normally a bit confusing for rabbits admin pages. > It runs perfectly on a debian system. Interesting. > ARGH! I can't use it there. > Now where is the difference? Weird thing i now use same java on both > servers. Since the problem seems to be with some gif images I have to ask if you have libgif and/or libungif installed on both systems? The gif libraries have had a few patent problems (ask unisys...) so they have not been installed by default. /robo |
|
From: Holger K. <hol...@gm...> - 2004-12-15 23:31:12
|
> Hmm, weird. > What version of rabbit do you use? 2.0.35 I tried 2.0.26 with the same result. I only changed the proxy port in the config file. Getting the config pages seems a bit difficult, but thats not the real problem. (IE 6.0 never asks for the password just fails with 407 proxy auth required, or it tells me / on that server couldn't be loaded. Mozilla works) The server in question has serveral ip adresses and dns names. Maybe that's confusing. It runs perfectly on a debian system. ARGH! I can't use it there. Now where is the difference? Weird thing i now use same java on both servers. Holger |
|
From: Robert O. <ro...@kh...> - 2004-12-15 23:02:11
|
Holger Krull wrote: > > Do you have any other resource that gives this problem? Or do I > > have to go to some page that has that image in it? > > The site www.gmx.de has the no recompression problem. > www.theinquirer.net gets recompressed perfectly. Ok, both of thoose sites gets compressed quite good for me. I can see that almost all images have been transformed. > > Do you have that image in your web browser cache? If you do rabbit > > will probably get a conditional request and answear with a "304 not > > modified". > > Cleared all caches, tried IE 6.0 firefox mozilla, konquerer. All have > problems. Hmm, weird. What version of rabbit do you use? /robo |
|
From: Robert O. <ro...@kh...> - 2004-12-15 23:00:37
|
Holger Krull wrote: > I just noticed another strange thing. If i use try > http://proxyhost:8080/ i have to enter the username and password > serveral times to get the status display. If i change to Connections > or config i have to reenter the username password serveral times > again. Rabbit does not listen on that port by default, so you obviously have changed some parts of the config. How do your config look? Do you have web proxy enabled for all web sites? including the one running rabbit? Since rabbit is a web proxy it understands proxy requests better. Rabbit tries to handle normal web server requests directed at it, but it fails quite often (mostly due to the fact that it tries to match the name given against one of the names of the machine). If you send a web server request to rabbit instead of a proxy request then rabbit will quite often ask everytime for username and password when you access its admin pages. > I deleted and reinstalled Rabbit, no change. Java version problem? Or > doesn't it like Suse 9.2? Will try a debian system next. I run on debian, but it should work fine on suse. /robo |
|
From: Holger K. <hol...@gm...> - 2004-12-15 23:00:18
|
> I do not get any expire-date parsing warning for that image either. > I get an image that is 6674 bytes long (same as you got). > > Do you have any other resource that gives this problem? Or do I have to go > to some page that has that image in it? The site www.gmx.de has the no recompression problem. www.theinquirer.net gets recompressed perfectly. > Do you have that image in your web browser cache? If you do rabbit will > probably get a conditional request and answear with a "304 not modified". Cleared all caches, tried IE 6.0 firefox mozilla, konquerer. All have problems. Weird thing? Holger |
|
From: Holger K. <hol...@gm...> - 2004-12-15 22:46:48
|
I just noticed another strange thing. If i use try http://proxyhost:8080/ i have to enter the username and password serveral times to get the status display. If i change to Connections or config i have to reenter the username password serveral times again. I deleted and reinstalled Rabbit, no change. Java version problem? Or doesn't it like Suse 9.2? Will try a debian system next. Holger |
|
From: Robert O. <ro...@kh...> - 2004-12-15 21:55:39
|
Holger Krull wrote: > i found that while using rabbit on some sites the images don't get > recompressed. > ... 'http://images.gmx.net/images/bs/21000/skyscraper_wechsler_v2.gif' Ok, I tried that image and I get a low quality image when I access that image. I do not get any expire-date parsing warning for that image either. I get an image that is 6674 bytes long (same as you got). Do you have any other resource that gives this problem? Or do I have to go to some page that has that image in it? Do you have that image in your web browser cache? If you do rabbit will probably get a conditional request and answear with a "304 not modified". /robo |
1065 messages has been excluded from this view by a project administrator.
