Re: [squid-users] Squid on FreeBSD poor performance

Firstly, drop cache_mem way way down to something like 128 or 256meg.
It doesn't need to be that big and that "can't allocate!" is FreeBSD
saying "no!" to memory allocations after your process grows past a certain
size.

200 req/sec shouldn't be making it cry, 800 req/sec should be.
I do think you've got way, way too many cache dirs though.
I'd collapse that to one per disk rather than lots of 8 gig
cachedirs.

FreeBSD-6 and FreeBSD-7 can use aufs instead of diskd. Try that maybe.

The dnsSubmit() messages are probably an indication your DNS server
isn't fast enough.

Other than that, its hard to tell whats going on without historical statistics.
Thats why I suggest setting up any kind of statistics graphing to all of my
clients and those that have thank me for it.

Adrian

On Wed, Mar 05, 2008, joejoe wrote:
> Hi!
>
> Today i'm running squid 2.6 stable18 on FreeBSD 6.3 in a Acer R710 Server.
> Client have 3000 users
>
> But performance is poor.
> Consults has the method to be allowed to improve?
>
> My hardware is:
>
> CPU: Intel(R) Xeon(TM) CPU 3.20GHz *2
> RAM: 4G
> HD: SEAGATE ST373207LC * 4 (320M/s 10000 RPM)
>
>
> #uname -a
> proxy4 -joejoe-:uname -a
> FreeBSD proxy4.tw 6.3-PRERELEASE FreeBSD 6.3-PRERELEASE #1: Sat Nov 17
> 11:45:41 CST 2007
> root@proxy4.tw:/usr/obj/usr/src/sys/PROXY i386
>
>
> #mount
> /dev/da0s1a on / (ufs, local)
> devfs on /dev (devfs, local)
> /dev/da0s1f on /squidlogs (ufs, local, soft-updates)
> /dev/da0s1d on /usr (ufs, local, soft-updates)
> /dev/da0s1e on /var (ufs, local, soft-updates)
> /dev/da0s1g on /tmp (ufs, local, soft-updates)
> /dev/da1s1d on /bcache1 (ufs, local, noatime, soft-updates)
> /dev/da1s1e on /bcache2 (ufs, local, noatime, soft-updates)
> /dev/da1s2d on /bcache3 (ufs, local, noatime, soft-updates)
> /dev/da1s2e on /bcache4 (ufs, local, noatime, soft-updates)
> /dev/da1s3d on /bcache5 (ufs, local, noatime, soft-updates)
> /dev/da1s3e on /bcache6 (ufs, local, noatime, soft-updates)
> /dev/da1s4d on /bcache7 (ufs, local, noatime, soft-updates)
> /dev/da1s4e on /bcache8 (ufs, local, noatime, soft-updates)
> /dev/da2s1d on /ccache1 (ufs, local, noatime, soft-updates)
> /dev/da2s1e on /ccache2 (ufs, local, noatime, soft-updates)
> /dev/da2s2d on /ccache3 (ufs, local, noatime, soft-updates)
> /dev/da2s2e on /ccache4 (ufs, local, noatime, soft-updates)
> /dev/da2s3d on /ccache5 (ufs, local, noatime, soft-updates)
> /dev/da2s3e on /ccache6 (ufs, local, noatime, soft-updates)
> /dev/da2s4d on /ccache7 (ufs, local, noatime, soft-updates)
> /dev/da2s4e on /ccache8 (ufs, local, noatime, soft-updates)
> /dev/da3s1d on /dcache1 (ufs, local, noatime, soft-updates)
> /dev/da3s1e on /dcache2 (ufs, local, noatime, soft-updates)
> /dev/da3s2d on /dcache3 (ufs, local, noatime, soft-updates)
> /dev/da3s2e on /dcache4 (ufs, local, noatime, soft-updates)
> /dev/da3s3d on /dcache5 (ufs, local, noatime, soft-updates)
> /dev/da3s3e on /dcache6 (ufs, local, noatime, soft-updates)
> /dev/da3s4d on /dcache7 (ufs, local, noatime, soft-updates)
> /dev/da3s4e on /dcache8 (ufs, local, noatime, soft-updates)
>
>
> #df -h
> Filesystem Size Used Avail Capacity Mounted on
> /dev/da0s1a 9.7G 70M 8.8G 1% /
> devfs 1.0K 1.0K 0B 100% /dev
> /dev/da0s1f 10G 1.2G 8.1G 13% /squidlogs
> /dev/da0s1d 19G 10G 7.8G 56% /usr
> /dev/da0s1e 19G 103M 18G 1% /var
> /dev/da0s1g 5.8G 92K 5.4G 0% /tmp
> /dev/da1s1d 7.8G 6.0G 1.2G 84% /bcache1
> /dev/da1s1e 7.8G 6.0G 1.1G 84% /bcache2
> /dev/da1s2d 7.8G 6.0G 1.1G 84% /bcache3
> /dev/da1s2e 7.8G 6.0G 1.1G 84% /bcache4
> /dev/da1s3d 7.8G 6.0G 1.1G 84% /bcache5
> /dev/da1s3e 7.8G 6.0G 1.2G 84% /bcache6
> /dev/da1s4d 7.8G 6.0G 1.1G 84% /bcache7
> /dev/da1s4e 7.8G 6.0G 1.1G 84% /bcache8
> /dev/da2s1d 7.8G 6.0G 1.2G 84% /ccache1
> /dev/da2s1e 7.8G 6.0G 1.1G 84% /ccache2
> /dev/da2s2d 7.8G 6.0G 1.1G 84% /ccache3
> /dev/da2s2e 7.8G 6.0G 1.1G 84% /ccache4
> /dev/da2s3d 7.8G 6.0G 1.2G 84% /ccache5
> /dev/da2s3e 7.8G 6.0G 1.1G 84% /ccache6
> /dev/da2s4d 7.8G 6.0G 1.2G 84% /ccache7
> /dev/da2s4e 7.8G 6.0G 1.2G 84% /ccache8
> /dev/da3s1d 7.8G 6.0G 1.1G 84% /dcache1
> /dev/da3s1e 7.8G 6.0G 1.1G 84% /dcache2
> /dev/da3s2d 7.8G 6.0G 1.1G 84% /dcache3
> /dev/da3s2e 7.8G 6.0G 1.2G 84% /dcache4
> /dev/da3s3d 7.8G 6.0G 1.1G 84% /dcache5
> /dev/da3s3e 7.8G 6.0G 1.2G 84% /dcache6
> /dev/da3s4d 7.8G 6.0G 1.1G 84% /dcache7
> /dev/da3s4e 7.8G 6.0G 1.2G 84% /dcache8
>
>
> My kernel was compiled in the day before yesterday with some tunning
> options:
>
> Commented:
>
> options PAE
> options MAXFILES=16384
>
> In /boot/loader.conf file :
>
> kern.ipc.msgmnb=16384 # Origin: 2048
> kern.ipc.msgmni=96 # Origin: 40
> kern.ipc.msgseg=2048 # Origin: 2048
> kern.ipc.msgssz=64 # Origin: 8
> kern.ipc.msgtql=5120 # Origin: 40
> kern.geom.debugflags=16
> machdep.hyperthreading_allowed=1
>
> kern.maxdsiz="2147483648" # 2GB
> kern.dfldsiz="2147483648" # 2GB
> kern.maxssiz="268435456" # 256MB
>
> kern.ipc.maxsockets="4008"
> kern.ipc.nmbclusters="32768"
> kern.ipc.nmbufs="65535"
> kern.ipc.nsfbufs="2496"
> net.inet.tcp.tcbhashsize="2048"
>
> In /etc/sysctl.conf file :
>
> net.inet.ip.portrange.first=20000 # Origin: 49152
> net.inet.tcp.blackhole=2
> net.inet.udp.blackhole=1
> net.inet.tcp.recvspace=65535
> net.inet.tcp.sendspace=65535
> kern.ipc.somaxconn=32768
> kern.ipc.maxsockbuf=2097152
> net.inet.ip.redirect=0
> net.inet.icmp.icmplim=100
> net.inet.udp.maxdgram=65535
> net.local.stream.sendspace=65535
> net.local.stream.recvspace=65535
> kern.maxfiles=65536
> kern.maxfilesperproc=65536
> net.inet.tcp.msl=7500
> kern.securelevel=0
>
>
> In squid.conf file :
>
> # SCSI Channel B HD * 8
>
> cache_dir diskd /bcache1 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /bcache2 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /bcache3 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /bcache4 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /bcache5 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /bcache6 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /bcache7 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /bcache8 7500 32 64 Q1=72 Q2=64
>
> # SCSI Channel C HD * 8
> cache_dir diskd /ccache1 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /ccache2 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /ccache3 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /ccache4 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /ccache5 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /ccache6 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /ccache7 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /ccache8 7500 32 64 Q1=72 Q2=64
>
> # SCSI Channel D HD * 8
> cache_dir diskd /dcache1 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /dcache2 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /dcache3 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /dcache4 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /dcache5 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /dcache6 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /dcache7 7500 32 64 Q1=72 Q2=64
> cache_dir diskd /dcache8 7500 32 64 Q1=72 Q2=64
>
> cache_mem 1536 MB
>
> cache_swap_low 80
> cache_swap_high 90
>
> maximum_object_size 16 MB
> minimum_object_size 0 KB
> maximum_object_size_in_memory 500 KB
>
> ipcache_size 32768
> ipcache_low 85
> ipcache_high 95
>
> fqdncache_size 32768
>
> cache_replacement_policy lru
>
> memory_replacement_policy lru
>
> dns_children 32
>
> request_body_max_size 100 MB
>
> half_closed_clients off
>
> store_avg_object_size 32 KB
>
>
> # top -d 1
> last pid: 60798; load averages: 1.02, 0.99, 0.86 up 2+06:36:15
> 11:09:01
> 100 processes: 2 running, 98 sleeping
> CPU states: % user, % nice, % system, % interrupt, %
> idle
> Mem: 1876M Active, 1750M Inact, 204M Wired, 165M Cache, 112M Buf, 5660K Free
> Swap: 8065M Total, 104K Used, 8065M Free
>
> PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND
> 60456 squid 1 105 0 1839M 1834M CPU2 2 31:06 83.89% squid
> 60490 squid 1 -4 0 1888K 988K msgwai 2 0:03 0.00%
> diskd-daemon
> 60498 squid 1 -4 0 1888K 1084K msgwai 2 0:03 0.00%
> diskd-daemon
> 60502 squid 1 -4 0 1888K 980K msgwai 0 0:03 0.00%
> diskd-daemon
> 60509 squid 1 -4 0 1888K 964K msgwai 0 0:02 0.00%
> diskd-daemon
> 60506 squid 1 -4 0 1888K 1004K msgwai 1 0:02 0.00%
> diskd-daemon
> ........
>
>
>
> #squid -v
>
> Squid Cache: Version 2.6.STABLE18
> configure options:
> '--prefix=/squid'
> '--enable-gnuregex'
> '--with-aufs-threads=320'
> '--with-pthreads'
> '--with-aio'
> '--with-dl'
> '--enable-async-io'
> '--enable-storeio=ufs,diskd,aufs,null'
> '--enable-removal-policies=heap lru'
> '--disable-wccp'
> '--enable-kill-parent-hack'
> '--enable-cachemgr-hostname=proxy4.tw'
> '--enable-cache-digests'
> '--enable-default-err-language=Traditional_Chinese'
> '--enable-err-languages=Traditional_Chinese'
> '--enable-poll'
> '--disable-ident-lookups'
> '--disable-internal-dns'
> '--disable-hostname-checks'
> '--enable-underscores'
> '--enable-stacktraces'
> '--enable-snmp'
> 'CFLAGS=-O2 -mcpu=pentium4 -march=pentium4 -mmmx -msse3 -m3dnow'
>
> #squidclient mgr:info
> HTTP/1.0 200 OK
> Server: squid/2.6.STABLE18
> Date: Wed, 05 Mar 2008 03:11:22 GMT
> Content-Type: text/plain
> Expires: Wed, 05 Mar 2008 03:11:22 GMT
> Last-Modified: Wed, 05 Mar 2008 03:11:22 GMT
> X-Cache: MISS from proxy4.tw
> X-Cache-Lookup: MISS from proxy4.tw:8080
> Via: 1.0 proxy4.tw:8080 (squid/2.6.STABLE18)
> Proxy-Connection: close
>
> Squid Object Cache: Version 2.6.STABLE18
> Start Time: Wed, 05 Mar 2008 02:12:49 GMT
> Current Time: Wed, 05 Mar 2008 03:11:22 GMT
> Connection information for squid:
> Number of clients accessing cache: 845
> Number of HTTP requests received: 437974
> Number of ICP messages received: 0
> Number of ICP messages sent: 0
> Number of queued ICP replies: 0
> Request failure ratio: 0.00
> Average HTTP requests per minute since start: 7480.9
> Average ICP messages per minute since start: 0.0
> Select loop called: 3341104 times, 1.051 ms avg
> Cache information for squid:
> Request Hit Ratios: 5min: 27.0%, 60min: 19.7%
> Byte Hit Ratios: 5min: 13.2%, 60min: 7.2%
> Request Memory Hit Ratios: 5min: 19.0%, 60min: 23.1%
> Request Disk Hit Ratios: 5min: 23.3%, 60min: 17.7%
> Storage Swap size: 147455723 KB
> Storage Mem size: 687156 KB
> Mean Object Size: 30.55 KB
> Requests given to unlinkd: 0
> Median Service Times (seconds) 5 min 60 min:
> HTTP Requests (All): 0.01387 0.09736
> Cache Misses: 0.06286 0.16775
> Cache Hits: 0.00091 0.00379
> Near Hits: 0.03066 0.04519
> Not-Modified Replies: 0.00000 0.00091
> DNS Lookups: 0.00190 0.00372
> ICP Queries: 0.00000 0.00000
> Resource usage for squid:
> UP Time: 3512.725 seconds
> CPU Time: 1993.014 seconds
> CPU Usage: 56.74%
> CPU Usage, 5 minute avg: 88.82%
> CPU Usage, 60 minute avg: 56.51%
> Process Data Segment Size via sbrk(): 1927696 KB
> Maximum Resident Size: 1935028 KB
> Page faults with physical i/o: 4
> Memory accounted for:
> Total accounted: 1087998 KB
> memPoolAlloc calls: 80133034
> memPoolFree calls: 64600088
> File descriptor usage for squid:
> Maximum number of file descriptors: 32768
> Largest file desc currently in use: 1228
> Number of file desc currently in use: 1118
> Files queued for open: 0
> Available number of file descriptors: 31650
> Reserved number of file descriptors: 100
> Store Disk files open: 0
> IO loop method: poll
> Internal Data Structures:
> 4828174 StoreEntries
> 40667 StoreEntries with MemObjects
> 40574 Hot Object Cache Items
> 4826951 on-disk objects
>
>
>
> #squidclient gr:5min | grep client
>
> client_http.requests = 202.585409/sec
> client_http.hits = 50.436354/sec
> client_http.errors = 0.006667/sec
> client_http.kbytes_in = 181.338874/sec
> client_http.kbytes_out = 2728.809730/sec
> client_http.all_median_svc_time = 0.012346 seconds
> client_http.miss_median_svc_time = 0.032412 seconds
> client_http.nm_median_svc_time = 0.000000 seconds
> client_http.nh_median_svc_time = 0.028993 seconds
> client_http.hit_median_svc_time = 0.001017 seconds
>
> cache.log frequently show messages:
>
> EX:
> 2008/03/05 06:18:10| Starting Squid Cache version 2.6.STABLE18 for
> i386-unknown-freebsd6.3...
> 2008/03/05 06:18:10| Process ID 59372
> 2008/03/05 06:18:10| With 32768 file descriptors available
> 2008/03/05 06:18:10| Using poll for the IO loop
> 2008/03/05 06:18:10| Performing DNS Tests...
> 2008/03/05 06:18:10| Successful DNS name lookup tests...
> 2008/03/05 06:18:10| helperOpenServers: Starting 32 'dnsserver' processes
> 2008/03/05 06:18:10| Unlinkd pipe opened on FD 41
> 2008/03/05 06:18:10| Swap maxSize 184320000 KB, estimated 5760000 objects
> 2008/03/05 06:18:10| Target number of buckets: 288000
> 2008/03/05 06:18:10| Using 524288 Store buckets
> 2008/03/05 06:18:10| Max Mem size: 1572864 KB
> 2008/03/05 06:18:10| Max Swap size: 184320000 KB
> 2008/03/05 06:18:10| Local cache digest enabled; rebuild/rewrite every
> 3600/3600 sec
> 2008/03/05 06:18:10| Store logging disabled
> 2008/03/05 06:18:10| Rebuilding storage in /bcache1 (CLEAN)
> 2008/03/05 06:18:10| Rebuilding storage in /bcache2 (CLEAN)
> 2008/03/05 06:18:10| Rebuilding storage in /bcache3 (CLEAN)
> 2008/03/05 06:18:10| Rebuilding storage in /bcache4 (CLEAN)
> 2008/03/05 06:18:10| Rebuilding storage in /bcache5 (CLEAN)
> ......
> 2008/03/05 06:18:46| Store rebuilding is 2.0% complete
> 2008/03/05 06:19:19| Store rebuilding is 2.1% complete
> 2008/03/05 06:19:49| Store rebuilding is 4.1% complete
> 2008/03/05 06:20:20| Store rebuilding is 6.2% complete
> 2008/03/05 06:20:49| Store rebuilding is 8.2% complete
> 2008/03/05 06:38:01| Store rebuilding is 98.1% complete
> 2008/03/05 06:38:02| Done reading /dcache7 swaplog (205231 entries)
> 2008/03/05 06:38:03| Done reading /ccache1 swaplog (206124 entries)
> 2008/03/05 06:38:03| Done reading /dcache4 swaplog (206889 entries)
> 2008/03/05 06:38:04| Done reading /dcache2 swaplog (207196 entries)
> 2008/03/05 06:38:05| Done reading /bcache3 swaplog (209643 entries)
> 2008/03/05 06:38:06| Done reading /dcache3 swaplog (211186 entries)
> 2008/03/05 06:38:06| Done reading /dcache1 swaplog (213883 entries)
> 2008/03/05 06:38:06| Finished rebuilding storage from disk.
> 2008/03/05 06:38:06| 4838833 Entries scanned
> 2008/03/05 06:38:06| 0 Invalid entries.
> 2008/03/05 06:38:06| 0 With invalid flags.
> 2008/03/05 06:38:06| 4838713 Objects loaded.
> 2008/03/05 06:38:06| 0 Objects expired.
> 2008/03/05 06:38:06| 0 Objects cancelled.
> 2008/03/05 06:38:06| 120 Duplicate URLs purged.
> 2008/03/05 06:38:06| 0 Swapfile clashes avoided.
> 2008/03/05 06:38:06| Took 1195.8 seconds (4046.3 objects/sec).
> 2008/03/05 06:38:06| Beginning Validation Procedure
> 2008/03/05 06:38:06| 262144 Entries Validated so far.
> 2008/03/05 06:38:06| 524288 Entries Validated so far.
> ......
> 2008/03/05 06:38:08| Completed Validation Procedure
> 2008/03/05 06:38:08| Validated 4838713 Entries
> 2008/03/05 06:38:08| store_swap_size = 147455350k
> 2008/03/05 06:38:11| storeLateRelease: released 142 objects
>
> 2008/03/05 06:58:34| httpReadReply: Excess data from "GET
> http://sugg.search.yahoo.com/sg/
> ?output=fxjsonp&nresults=10&command=www"
> .......
> 2008/03/05 07:23:51| httpAccept: FD 112: accept failure: (53) Software
> caused connection abort
> 2008/03/05 07:27:02| httpReadReply: Excess data from "GET
> http://webcs.msg.yahoo.com/crossdomain.xml"
> ........
> 2008/03/05 08:51:10| httpAccept: FD 112: accept failure: (53) Software
> caused connection abort
> 2008/03/05 08:51:10| httpAccept: FD 112: accept failure: (53) Software
> caused connection abort
> 2008/03/05 08:51:10| httpAccept: FD 112: accept failure: (53) Software
> caused connection abort
> 2008/03/05 08:51:10| httpAccept: FD 112: accept failure: (53) Software
> caused connection abort
> 2008/03/05 08:51:10| httpAccept: FD 112: accept failure: (53) Software
> caused connection abort
> .......
> 2008/03/05 09:17:07| parseHttpRequest: Unsupported method '^E^ACONNECT'
> 2008/03/05 09:17:07| clientReadRequest: FD 860 (140.1.144.197:2664) Invalid
> Request
> 2008/03/05 09:19:56| urlParse: Illegal character in hostname
> 'mlbglobal,.112.2o7.net'
> 2008/03/05 09:21:00| WARNING: 10 swapin MD5 mismatches
> ......
> 2008/03/05 09:35:30| WARNING: All dnsserver processes are busy.
> 2008/03/05 09:35:30| WARNING: up to 32 pending requests queued
> ......
> 2008/03/05 10:05:07| parseHttpRequest: Unsupported method '^E^ACONNECT'
> 2008/03/05 10:05:07| clientReadRequest: FD 446 (140.1.144.197:3222) Invalid
> Request
> 2008/03/05 10:05:13| WARNING: All dnsserver processes are busy.
> 2008/03/05 10:05:13| WARNING: up to 50 pending requests queued
> 2008/03/05 10:05:13| dnsSubmit: queue overload, rejecting l.yimg.com
> 2008/03/05 10:05:13| dnsSubmit: queue overload, rejecting l.yimg.com
> 2008/03/05 10:05:13| dnsSubmit: queue overload, rejecting l.yimg.com
> 2008/03/05 10:05:13| dnsSubmit: queue overload, rejecting l.yimg.com
> 2008/03/05 10:05:13| dnsSubmit: queue overload, rejecting l.yimg.com
> 2008/03/05 10:12:45| httpReadReply: Excess data from "GET http://38.114.207.
> 9/4939beeaad8cea7f780288f9ecbb670d5
> FATAL: xcalloc: Unable to allocate 1 blocks of 4108 bytes!
>
> Squid Cache (Version 2.6.STABLE18): Terminated abnormally.
> CPU Usage: 3700.403 seconds = 444.883 user + 3255.521 sys
> Maximum Resident Size: 2102284 KB
> Page faults with physical i/o: 0
> 2008/03/05 10:12:45| ctx: enter level 0:
> 'http://www.schprs.edu.tw/images/layout_index_13.gif'
> 2008/03/05 10:12:45| storeDirWriteCleanLogs: Starting...
> 2008/03/05 10:12:49| Starting Squid Cache version 2.6.STABLE18 for
> i386-unknown-freebsd6.3...
> 2008/03/05 10:12:49| Process ID 60456
> 2008/03/05 10:12:49| With 32768 file descriptors available
> 2008/03/05 10:12:49| Using poll for the IO loop
> 2008/03/05 10:12:49| Performing DNS Tests...
> 2008/03/05 10:12:49| Successful DNS name lookup tests...
> 2008/03/05 10:12:49| helperOpenServers: Starting 32 'dnsserver' processes
> 2008/03/05 10:12:49| Unlinkd pipe opened on FD 41
> 2008/03/05 10:12:49| Swap maxSize 184320000 KB, estimated 5760000 objects
> 2008/03/05 10:12:49| Target number of buckets: 288000
> 2008/03/05 10:12:49| Using 524288 Store buckets
> 2008/03/05 10:12:49| Max Mem size: 1572864 KB
> 2008/03/05 10:12:49| Max Swap size: 184320000 KB
> 2008/03/05 10:12:49| Local cache digest enabled; rebuild/rewrite every
> 3600/3600 sec
> 2008/03/05 10:12:49| Store logging disabled
> 2008/03/05 10:12:49| Rebuilding storage in /bcache1 (DIRTY)
> 2008/03/05 10:12:49| Rebuilding storage in /bcache2 (DIRTY)
> 2008/03/05 10:12:49| Rebuilding storage in /bcache3 (DIRTY)
> ........
> 2008/03/05 10:37:22| httpReadReply: Excess data from "GET http://pic.wretch.
> cc/icon/blog/melinda/20050607/comme
> 2008/03/05 10:37:22| 1048576 Entries Validated so far.
> 2008/03/05 10:37:23| 1310720 Entries Validated so far.
> 2008/03/05 10:37:23| 1572864 Entries Validated so far.
> 2008/03/05 10:37:23| 1835008 Entries Validated so far.
> 2008/03/05 10:37:23| 2097152 Entries Validated so far.
> 2008/03/05 10:37:24| 2359296 Entries Validated so far.
> 2008/03/05 10:37:24| 2621440 Entries Validated so far.
> 2008/03/05 10:37:24| 2883584 Entries Validated so far.
> 2008/03/05 10:37:25| 3145728 Entries Validated so far.
> 2008/03/05 10:37:25| 3407872 Entries Validated so far.
> 2008/03/05 10:37:25| 3670016 Entries Validated so far.
> 2008/03/05 10:37:26| 3932160 Entries Validated so far.
> 2008/03/05 10:37:26| 4194304 Entries Validated so far.
> 2008/03/05 10:37:26| 4456448 Entries Validated so far.
> 2008/03/05 10:37:26| 4718592 Entries Validated so far.
> 2008/03/05 10:37:26| httpAccept: FD 112: accept failure: (53) Software
> caused connection abort
> 2008/03/05 10:37:27| httpAccept: FD 112: accept failure: (53) Software
> caused connection abort
> 2008/03/05 10:37:27| Completed Validation Procedure
> 2008/03/05 10:37:27| Validated 4794164 Entries
> 2008/03/05 10:37:27| store_swap_size = 147191853k
>
>
> squid is always restart
> cache.log show messages
>
> FATAL: xmalloc: Unable to allocate 65535 bytes!
> or
> FATAL: xcalloc: Unable to allocate 1 blocks of 4108 bytes!
>
> Any suggestion?!?!
>
> tanx
>
>
> --

-- 
- Xenion - http://www.xenion.com.au/ - VPS Hosting - Commercial Squid Support -
- $25/pm entry-level VPSes w/ capped bandwidth charges available in WA -