Re: [squid-users] calling Henrick , Joe ,Adrian ...SOS pls help out. pls pls pls

From: khiz code <khizcode@dont-contact.us>
Date: Fri, 17 Aug 2001 21:35:35 -0700 (PDT)

Hi Joe
i did put the squid box under load for abt 5 mins and
came up with the same disappointing results
well DNS was responding quickly enuf
howver telnet localhost 3128 from the same box was
taking ages after just 2 mins under load
i am sending u the cachemgr stats which i could
collect immediately after putting off the load (
during load cachemgr too could not connect to squid)

to clarify i put forth the foll details
1> compaq prolinea 5500 dual pii xeon 550 Mhz/512 mb
ram/6 scsi disk of 9 GB each
  RAID 0 done on 4 of the drives and partitioned as
/cache using reiserfs for effective 34 GB
RAID 1 done on the remaining drives where linux is
installed
2> redhat 6.2 with kernel 2.2.19 and reiserfs 3.53
  scsi driver sym 53cxx compiled in . along with tlan
driver
3> cache box set up as transparent proxy with ipchains
target prot opt source
destination ports
ACCEPT all ------ localhost localhost
n/a
REDIRECT tcp ------ anywhere anywhere
            any -> www
=> 3128
ACCEPT all ------ anywhere anywhere
            n/a
Chain forward (policy ACCEPT):
Chain output (policy ACCEPT):
4> squid.conf
cache_mem 128 MB
maximum_object_size 40960 KB
cache_dir /cache 10000 30 25
dns_children 15
httpd_accel_host virtual
httpd_accel_port 80
httpd_accel_with_proxy on
httpd_accel_uses_host_header on
memory_pools_limit 70 MB
5> glibc-2.1.3-22

STATISTICS------
CAHE RUN TIME
Connection information for squid:
        Number of clients accessing cache: 40
        Number of HTTP requests received: 4168
        Number of ICP messages received: 0
        Number of ICP messages sent: 0
        Number of queued ICP replies: 0
        Request failure ratio: 0.00%
        HTTP requests per minute: 94.7
        ICP messages per minute: 0.0
        Select loop called: 187023 times, 14.120 ms avg
Cache information for squid:
        Request Hit Ratios: 5min: 10.1%, 60min: 6.3%
        Byte Hit Ratios: 5min: 23.0%, 60min: 20.0%
        Storage Swap size: 6182 KB
        Storage Mem size: 3004 KB
        Storage LRU Expiration Age: 365.00 days
        Mean Object Size: 8.63 KB
        Requests given to unlinkd: 21
Median Service Times (seconds) 5 min 60 min:
        HTTP Requests (All): 0.01648 0.02899
        Cache Misses: 0.61549 0.58309
        Cache Hits: 0.00598 0.00562
        Near Hits: 0.04776 0.04047
        Not-Modified Replies: 0.00678 0.00562
        DNS Lookups: 0.01210 0.01269
        ICP Queries: 0.00000 0.00000
Resource usage for squid:
        UP Time: 2640.817 seconds
        CPU Time: 76.890 seconds
        CPU Usage: 2.91%
        CPU Usage, 5 minute avg: 17.93%
        CPU Usage, 60 minute avg: 2.91%
        Maximum Resident Size: 0 KB
        Page faults with physical i/o: 334
Memory usage for squid via mallinfo():
        Total space in arena: 15266 KB
        Ordinary blocks: 14696 KB 537 blks
        Small blocks: 0 KB 0 blks
        Holding blocks: 2952 KB 2 blks
        Free Small blocks: 0 KB
        Free Ordinary blocks: 570 KB
        Total in use: 17648 KB 116%
        Total free: 570 KB 4%
Memory accounted for:
        Total accounted: 4126 KB
File descriptor usage for squid:
        Maximum number of file descriptors: 16384
        Largest file desc currently in use: 251
        Number of file desc currently in use: 83
        Files queued for open: 0
        Available number of file descriptors: 16301
        Reserved number of file descriptors: 100
        Store Disk files open: 1
Internal Data Structures:
           740 StoreEntries
           656 StoreEntries with MemObjects
           650 Hot Object Cache Items
           717 Filemap bits set
           716 on-disk objects
===============================================
ASYNC IO Counters:

open 753
close 1
cancel 8413
write 0
read 47
stat 0
unlink 1
check_callback 190484
queue 0
================================================
sample_start_time = 997626385.518401 (Sun, 12 Aug 2001
14:26:25 GMT)
sample_end_time = 997626685.573941 (Sun, 12 Aug 2001
14:31:25 GMT)
client_http.requests = 3.306055/sec
client_http.hits = 0.356601/sec
client_http.errors = 0.000000/sec
client_http.kbytes_in = 2.312905/sec
client_http.kbytes_out = 8.331791/sec
client_http.all_median_svc_time = 0.016481 seconds
client_http.miss_median_svc_time = 0.615488 seconds
client_http.nm_median_svc_time = 0.006779 seconds
client_http.nh_median_svc_time = 0.030657 seconds
client_http.hit_median_svc_time = 0.006779 seconds
server.all.requests = 1.563044/sec
server.all.errors = 0.000000/sec
server.all.kbytes_in = 6.195520/sec
server.all.kbytes_out = 0.833179/sec
server.http.requests = 1.563044/sec
server.http.errors = 0.000000/sec
server.http.kbytes_in = 6.195520/sec
server.http.kbytes_out = 0.833179/sec
server.ftp.requests = 0.000000/sec
server.ftp.errors = 0.000000/sec
server.ftp.kbytes_in = 0.000000/sec
server.ftp.kbytes_out = 0.000000/sec
server.other.requests = 0.000000/sec
server.other.errors = 0.000000/sec
server.other.kbytes_in = 0.000000/sec
server.other.kbytes_out = 0.000000/sec
icp.pkts_sent = 0.000000/sec
icp.pkts_recv = 0.000000/sec
icp.queries_sent = 0.000000/sec
icp.replies_sent = 0.000000/sec
icp.queries_recv = 0.000000/sec
icp.replies_recv = 0.000000/sec
icp.replies_queued = 0.000000/sec
icp.query_timeouts = 0.000000/sec
icp.kbytes_sent = 0.000000/sec
icp.kbytes_recv = 0.000000/sec
icp.q_kbytes_sent = 0.000000/sec
icp.r_kbytes_sent = 0.000000/sec
icp.q_kbytes_recv = 0.000000/sec
icp.r_kbytes_recv = 0.000000/sec
icp.query_median_svc_time = 0.000000 seconds
icp.reply_median_svc_time = 0.000000 seconds
dns.median_svc_time = 0.009036 seconds
unlink.requests = 0.019996/sec
page_faults = 0.000000/sec
select_loops = 74.712835/sec
select_fds = 28.151455/sec
average_select_fd_period = 0.003573/fd
median_select_fds = 0.000000
swap_files_cleaned = 0.000000/sec
aborted_requests = 0.156638/sec
syscalls.polls = 77.532313/sec
syscalls.disk.opens = 0.526569/sec
syscalls.disk.closes = 0.526569/sec
syscalls.disk.reads = 0.026662/sec
syscalls.disk.writes = 5.079060/sec
syscalls.disk.seeks = 0.000000/sec
syscalls.disk.unlinks = 0.000000/sec
syscalls.sock.accepts = 4.705795/sec
syscalls.sock.sockets = 0.876504/sec
syscalls.sock.connects = 0.876504/sec
syscalls.sock.binds = 0.000000/sec
syscalls.sock.closes = 3.655990/sec
syscalls.sock.reads = 13.270876/sec
syscalls.sock.writes = 10.364748/sec
syscalls.sock.recvfroms = 0.003333/sec
syscalls.sock.sendtos = 0.000000/sec
cpu_time = 51.140000 seconds
wall_time = 300.055540 seconds
cpu_usage = 17.043511%
=================================================
DNS stats
# FD # Requests Flags Time Offset Request
1 4 192 A 412.527 0 (none)
2 7 33 A 454.254 0 (none)
3 8 12 A 1107.534 0 (none)
4 9 2 A 1107.520 0 (none)
5 10 0 A 1194505.417 0 (none)
6 11 0 A 1194505.417 0 (none)
7 12 0 A 1194505.417 0 (none)
8 13 0 A 1194505.417 0 (none)
9 14 0 A 1194505.417 0 (none)
10 15 0 A 1194505.417 0 (none)
11 16 0 A 1194505.417 0 (none)
12 17 0 A 1194505.417 0 (none)
13 18 0 A 1194505.417 0 (none)
14 19 0 A 1194505.417 0 (none)
15 20 0 A 1194505.417 0 (none)

===========================================

hope this is enuf
shud i downgrade or upgrade the kernel to 2.4 and use
diskd instead of async-io

PLS HELP ME OUT..
ijust cannot let go of using squid on this lovely
piece of hardware
expecting all cooperation from squid gurus
rgds
khizcode
P.S : on this dual CPU machine when i telnet to it i
get
Red Hat Linux release 6.2 (Zoot)
Kernel 2.2.19 on a 2-processor i686
However 'top ' shows me
111 processes: 110 sleeping, 1 running, 0 zombie, 0
stopped
CPU states: 0.2% user, 0.6% system, 0.0% nice,
99.0% idle
no mention abt CPU0 and CPU 1 as is seen ????

--- Joe Cooper <joe@swelltech.com> wrote:
> Most Squid processes being in a sleeping state is
> normal except when
> under heavy load, particularly with that many
> threads (which you have
> too many of for your machine, it seems to me).
>
> You're going to have to tell us more. What is your
> cache.log telling
> you? Have you checked to insure DNS is responding
> quickly?
>
> What does cachemgr say about filedescriptor usage?
> (i.e. 'client
> mgr:info' what does the filedescriptor section tell
> you?)
>
> khiz code wrote:
>
> > HEllo all
> > welll ive been trying since the last week to get
> my
> > squid box to take some load
> > ive followed all the tips given in joe cooper's
> page
> > increased file descriptor to 16384
> > but even at less than 20 req/sec squid almost
> goes to
> > sleep
> > even telnetting from the same box to the squid
> port
> > takes abt a minute
> > even cachemgr is very slow
> > i dunno whats wrong
> > ive compiled squid with 55 threads and async-io on
> > redhat linux 6.2 kernel 2.2.19 with latest glibc
> > library
> > thrown 128MB out ofphysical mem of 512 MB for
> squid
> > yet it just goes to sleep does not respond !!!!1
> > top shows me
> > 120 processes: 119 sleeping, 1 running, 0 zombie,
> 0
> > stopped
> > i dunno what the problem is
> > pls help me out
> > i ve been mailing the list like mad but to no
> avail
> > PLs pls tell me what may be possibly worn
> > without load squid performs like a beauty
> > tearing my hair
> > khizcode
>
>
> --
> Joe Cooper <joe@swelltech.com>
> Affordable Web Caching Proxy
> Appliances
> http://www.swelltech.com
>

__________________________________________________
Do You Yahoo!?
Make international calls for as low as $.04/minute with Yahoo! Messenger
http://phonecard.yahoo.com/
Received on Fri Aug 17 2001 - 22:35:37 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:01:43 MST