Diagnostic Help

[Machsterdaemon]

I did see the www and non-www urls in firebug, but the redirection issue seems to have changed now for the better and the load time improved considerably when I browse your forums.

But the loading time needs much more improvement.

[Machsterdaemon]

Do you need having Gravatars enabled on your forums?

The page loading might get some improvements by disabling Gravatars.

[Gauravk]

Thanks for checking and notifying for improvement on www, but still redirection is there somewhere and I cant remember at what level I have done that in past.

About the gravatar, i have disabled it now for testing sake however I prefer such nice stuff if it doesn't cause too much delay.

Please find atest sql report below:

-------- General Statistics --------------------------------------------------
[--] Skipped version check for MySQLTuner script
[OK] Currently running supported MySQL version 5.5.37-cll
[OK] Operating on 32-bit architecture with less than 2GB RAM


-------- Storage Engine Statistics -------------------------------------------
[--] Status: +Archive -BDB -Federated +InnoDB -ISAM -NDBCluster
[--] Data in MyISAM tables: 122M (Tables: 895)
[--] Data in InnoDB tables: 330M (Tables: 3779)
[--] Data in PERFORMANCE_SCHEMA tables: 0B (Tables: 17)
[--] Data in MEMORY tables: 0B (Tables: 2)
[!!] Total fragmented tables: 244


-------- Security Recommendations  -------------------------------------------
[OK] All database users have passwords assigned


-------- Performance Metrics -------------------------------------------------
[--] Up for: 17h 29m 2s (673K q [10.695 qps], 36K conn, TX: 27B, RX: 126M)
[--] Reads / Writes: 54% / 46%
[--] Total buffers: 208.0M global + 10.2M per thread (50 max threads)
[OK] Maximum possible memory usage: 717.4M (35% of installed RAM)
[OK] Slow queries: 0% (347/673K)
[OK] Highest usage of available connections: 52% (26/50)
[OK] Key buffer size / total MyISAM indexes: 64.0M/41.0M
[OK] Key buffer hit rate: 96.1% (3M cached / 141K reads)
[OK] Query cache efficiency: 78.9% (378K cached / 479K selects)
[!!] Query cache prunes per day: 9260
[OK] Sorts requiring temporary tables: 0% (0 temp sorts / 9K sorts)
[OK] Temporary tables created on disk: 18% (6K on disk / 34K total)
[OK] Thread cache hit rate: 99% (26 created / 36K connections)
[!!] Table cache hit rate: 0% (1K open / 2M opened)
[OK] Open file limit used: 3% (724/20K)
[OK] Table locks acquired immediately: 99% (277K immediate / 278K locks)
[!!] InnoDB data size / buffer pool: 330.6M/32.0M


-------- Recommendations -----------------------------------------------------
General recommendations:
    Run OPTIMIZE TABLE to defragment tables for better performance
    MySQL started within last 24 hours - recommendations may be inaccurate
    Increase table_cache gradually to avoid file descriptor limits
Variables to adjust:
    query_cache_size (> 32M)
    table_cache (> 1024)
    innodb_buffer_pool_size (>= 330M)

[Gauravk]

How to kill this yellow wait....?

Any premium DNS option from godaddy can help in this case...?

[Aiwa]

When you see that long wait, check top and see what your %wa is.

[Dmacleo]

I'd take a guess at disc speed, stab in the dark though w/o actual data

[Grumpy]

How to kill this yellow wait....?

Any premium DNS option from godaddy can help in this case...?

yellow wait.jpg

I'd suggest you turn on debug in your acp.
System Settings > System > General Configuration > Debug level > [1:...] and save.

That'll make IPB's own measuring tools visible at the bottom of the page. It'll show you how long it took to generate a single page purely from server side. If it takes ~2s from just the server, it's time to optimize.

[Gauravk]

Hi Guys,

I found a tool at serverbear to run the script at my godaddy VPS thru SSH to find out how good/bad this server is?

Please check below and advice your opinion on below benchmark. Thanks.

# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #

Version 5.1.3 Based on the Byte Magazine Unix Benchmark

Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com


1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput 1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3

1 x File Copy 256 bufsize 500 maxblocks 1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3

1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10

1 x Process Creation 1 2 3

1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent) 1 2 3

1 x Shell Scripts (8 concurrent) 1 2 3

8 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10

8 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10

8 x Execl Throughput 1 2 3

8 x File Copy 1024 bufsize 2000 maxblocks 1 2 3

8 x File Copy 256 bufsize 500 maxblocks 1 2 3

8 x File Copy 4096 bufsize 8000 maxblocks 1 2 3

8 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10

8 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10

8 x Process Creation 1 2 3

8 x System Call Overhead 1 2 3 4 5 6 7 8 9 10

8 x Shell Scripts (1 concurrent) 1 2 3

8 x Shell Scripts (8 concurrent) 1 2 3

========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)

System: ******************
OS: GNU/Linux -- 2.6.18-028stab107.1 -- #1 SMP Wed Apr 17 19:10:55 MSD 2013
Machine: i686 (i386)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel® Xeon® CPU L5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel® Xeon® CPU L5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 2: Intel® Xeon® CPU L5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 3: Intel® Xeon® CPU L5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 4: Intel® Xeon® CPU L5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 5: Intel® Xeon® CPU L5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 6: Intel® Xeon® CPU L5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 7: Intel® Xeon® CPU L5506 @ 2.13GHz (4256.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
17:31:31 up 2 days, 10:42, 1 user, load average: 0.70, 1.26, 1.42; runlevel

------------------------------------------------------------------------
Benchmark Run: Thu Jul 10 2014 17:31:31 - 18:00:57
8 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 7064795.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 1906.0 MWIPS (10.2 s, 7 samples)
Execl Throughput 2549.1 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 270067.4 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 79918.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 452922.7 KBps (30.0 s, 2 samples)
Pipe Throughput 534725.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 164749.3 lps (10.0 s, 7 samples)
Process Creation 7436.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1817.3 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 354.1 lpm (60.1 s, 2 samples)
System Call Overhead 440647.3 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 7064795.3 605.4
Double-Precision Whetstone 55.0 1906.0 346.5
Execl Throughput 43.0 2549.1 592.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 270067.4 682.0
File Copy 256 bufsize 500 maxblocks 1655.0 79918.4 482.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 452922.7 780.9
Pipe Throughput 12440.0 534725.6 429.8
Pipe-based Context Switching 4000.0 164749.3 411.9
Process Creation 126.0 7436.2 590.2
Shell Scripts (1 concurrent) 42.4 1817.3 428.6
Shell Scripts (8 concurrent) 6.0 354.1 590.1
System Call Overhead 15000.0 440647.3 293.8
========
System Benchmarks Index Score 500.7

------------------------------------------------------------------------
Benchmark Run: Thu Jul 10 2014 18:00:57 - 18:40:43
8 CPUs in system; running 8 parallel copies of tests

Dhrystone 2 using register variables 12228592.8 lps (10.1 s, 7 samples)
Double-Precision Whetstone 14682.9 MWIPS (10.0 s, 7 samples)
Execl Throughput 4226.2 lps (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 263667.8 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 85035.7 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 595678.1 KBps (30.0 s, 2 samples)
Pipe Throughput 1298340.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching 343226.8 lps (10.1 s, 7 samples)
Process Creation 7354.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 3445.1 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 566.6 lpm (60.5 s, 2 samples)
System Call Overhead 865797.3 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 12228592.8 1047.9
Double-Precision Whetstone 55.0 14682.9 2669.6
Execl Throughput 43.0 4226.2 982.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 263667.8 665.8
File Copy 256 bufsize 500 maxblocks 1655.0 85035.7 513.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 595678.1 1027.0
Pipe Throughput 12440.0 1298340.8 1043.7
Pipe-based Context Switching 4000.0 343226.8 858.1
Process Creation 126.0 7354.1 583.7
Shell Scripts (1 concurrent) 42.4 3445.1 812.5
Shell Scripts (8 concurrent) 6.0 566.6 944.3
System Call Overhead 15000.0 865797.3 577.2
========
System Benchmarks Index Score 882.7

IOPSI/O

IOPSI/O Pingsioping -c 10
request=1 time=3.4 ms
request=2 time=4.0 ms
request=3 time=2.9 ms
request=4 time=4.0 ms
request=5 time=0.1 ms
request=6 time=4.0 ms
request=7 time=0.4 ms
request=8 time=2.0 ms
request=9 time=4.0 ms
request=10 time=4.0 ms

10 requests completed in 9039.6 ms, 347 iops, 1.4 mb/sI/O Seek Test (No Cache)ioping -RD
666 iops, 2.6 mb/s
min/avg/max/mdev = 0.1/1.5/122.3/4.4 msI/O Reads - Sequentialioping -RL
63 iops, 15.8 mb/s
min/avg/max/mdev = 0.7/15.9/608.0/47.0 msI/O Reads - Cachedioping -RC
153275 iops, 598.7 mb/s
min/avg/max/mdev = 0.0/0.0/0.1/0.0 msDDdd if=/dev/zero of=sb-io-test bs=1M count=1k conv=fdatasync
12.6679 seconds, 84.8 MB/sdd if=/dev/zero of=sb-io-test bs=64k count=16k conv=fdatasync
13.5427 seconds, 79.3 MB/sdd if=/dev/zero of=sb-io-test bs=1M count=1k oflag=dsync
15.6979 seconds, 68.4 MB/sdd if=/dev/zero of=sb-io-test bs=64k count=16k oflag=dsync
132.309 seconds, 8.1 MB/sFIORead IOPS233.0Read Bandwidth955 KB/secondWrite IOPS1419.0Write Bandwidth5.6 MB/secondRaw FIO OutputFIO random reads:
randomreads: (g=0): rw=randread, bs=4K-4K/4K-4K, ioengine=libaio, iodepth=64
fio-2.0.9
Starting 1 process
randomreads: Laying out IO file(s) (1 file(s) / 1024MB)

randomreads: (groupid=0, jobs=1): err= 0: pid=22261: Thu Jul 10 17:21:11 2014
read : io=1024.3MB, bw=955866 B/s, iops=233 , runt=1123588msec
cpu : usr=0.06%, sys=0.27%, ctx=166348, majf=0, minf=0
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=100.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0%
issued : total=r=262207/w=0/d=0, short=r=0/w=0/d=0

Run status group 0 (all jobs):
READ: io=1024.3MB, aggrb=933KB/s, minb=933KB/s, maxb=933KB/s, mint=1123588msec, maxt=1123588msec
Done

FIO random writes:
randomwrites: (g=0): rw=randwrite, bs=4K-4K/4K-4K, ioengine=libaio, iodepth=64
fio-2.0.9
Starting 1 process

randomwrites: (groupid=0, jobs=1): err= 0: pid=31940: Thu Jul 10 17:24:16 2014
write: io=1024.3MB, bw=5677.7KB/s, iops=1419 , runt=184729msec
cpu : usr=0.08%, sys=1.17%, ctx=2031, majf=0, minf=0
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=100.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0%
issued : total=r=0/w=0/d=262207, short=r=0/w=0/d=0

Run status group 0 (all jobs):
WRITE: io=1024.3MB, aggrb=5677KB/s, minb=5677KB/s, maxb=5677KB/s, mint=184729msec, maxt=184729msec
Done

Download Speed

Benchmarking download from Cachefly (http://cachefly.cachefly.net/100mb.test)
Got 11.0MB/s
Benchmarking download from Linode, Atlanta, GA, USA (http://speedtest.atlanta.lin
Got 5.29MB/s
Benchmarking download from Linode, Dallas, TX, USA (http://speedtest.dallas.linod
Got 6.55MB/s
Benchmarking download from Linode, Tokyo, JP (http://speedtest.tokyo.linode.com/1
Got 11.5MB/s
Benchmarking download from Linode, London, UK (http://speedtest.london.linode.com
Got 5.02MB/s
Benchmarking download from OVH, Paris, France (http://proof.ovh.net/files/100Mio.
Got 2.76MB/s
Benchmarking download from SmartDC, Rotterdam, Netherlands (http://mirror.i3d.net
Got 4.69MB/s
Benchmarking download from Hetzner, Nuernberg, Germany (http://hetzner.de/100MB.i
Got 1.92MB/s
Benchmarking download from iiNet, Perth, WA, Australia (http://ftp.iinet.net.au/t
Got 10.7MB/s
Benchmarking download from MammothVPS, Sydney, Australia (http://www.mammothvpscu
Got 1.08MB/s
Benchmarking download from Leaseweb, Haarlem, NL (http://mirror.nl.leaseweb.net/s
Got 4.10MB/s
Benchmarking download from Leaseweb, Manassas, VA, USA (http://mirror.us.leaseweb
Got 4.16MB/s
Benchmarking download from Softlayer, Singapore (http://speedtest.sng01.softlayer
Got 48.3MB/s
Benchmarking download from Softlayer, Seattle, WA, USA (http://speedtest.sea01.so
Got 4.15MB/s
Benchmarking download from Softlayer, San Jose, CA, USA (http://speedtest.sjc01.s
Got 4.05MB/s
Benchmarking download from Softlayer, Washington, DC, USA (http://speedtest.wdc01
Got 4.48MB/s

[Grumpy]

Unix bench is meant for dedicated servers and dedicated only that is not in production. It's made to test hardware, not soft caps. Running it in VPS either creates a lock up/slow down for all the users in that container and/or it produces false results from failing to interact with the intended interface. Same can be said for most benches. So, you really shouldn't run benches on a vps. If the provider failed to put caps very well, it would have crashed every site/app running on that vps. Some vps providers will warn/terminate accounts that run benchmarks for resource abuse, and those are actually the good ones because it means they monitor abusive neighbors.

Simple ioping without any additional commands is okay on a vps because it's very lightweight. But heavy read/writes will create above problem as well.

request=1 time=3.4 ms
request=2 time=4.0 ms
request=3 time=2.9 ms
request=4 time=4.0 ms
request=5 time=0.1 ms
request=6 time=4.0 ms
request=7 time=0.4 ms
request=8 time=2.0 ms
request=9 time=4.0 ms
request=10 time=4.0 ms

^ That is bad. Just from your result, it shows that it CAN do 0.1ms, yet does ~4 (40x slower) most of the time. An unused disk will produce consistent results.

63 iops > About as good as single low~mid disk consumer grade harddrive in iops.

The download speed tests the wrong direction of the pipe. You care about it's upstream, not it's downstream. These download files are provided by those datacenters because people want to test that datacenter's capability w.r.t. yours. But it does show a stark difference once it gets out of singapore, which is interesting. It looks like the connection to outside of singapore is 100mbps line/switch since it appears to be capping out at ~11MB/s (which is 100mbps) while it's 1gbps intra-nationally which would be pretty bad. But it's also possible it's just coincidence in data. And once it leaves asia, it gets degraded even more, which is expected since further... If you actually bought a dedi from softlayer (a high end dedi provider) in singapore and tested vs softlayer san jose, you'd expect a max speed. So, not that it's always expected, but just commonly expected. But again, this tests the wrong direction of the pipe, so there's no reason for your provider to even do as much optimization in this direction.

[Gauravk]

Hi All,

I like to thanks everyone here, who helped in identifying the server issue after the migration. After lot of research, I finally switched the server few days back and got the load time of 2 seconds.

As I am new to IPB, please check and advice if any further optimization or enhancement can be done at http://carnity.com

Feel free to comment on the community overall look and feel and architecture, as now I have time to give some love to cosmetic stuff.