Solaris 9, V210, bge0 network performance issue

Setup: Solaris 9, V210, single network (bge0).
ndd shows: 100FD
LAN Switch set to 100FD
Network performance bad, in fact very bad. System shows upwards of 19% collisions with netstat -- odd since FD should have no collisions. Network folks have checked switch settings and they are 100FD. I have checked more than once that ndd returns 100FD.
I am not sure what to check next. Thanks for any help you can provide in what to check next or pointing me to something I have missed.
Thanks - Wes

Means your system and the switch aren't playing nicely.
Anytime a Sun internal is showing collisions (don't care what ndd says), it means the interface is probably flip flopping between FD and HD (causing your collisions). The flip flops are causing performance to be MUCH worse then just the collisions ever would.
You might try a new factory made CAT5 network cable.
Then if that doesn't work, you should try using ndd to force the bge interface to 100FD and forcing the switch to 100FD.

Similar Messages

X6450 network performance issue

We are seeing a network performance issue using the Blade 6000 chassis with two (10x 1 Gbps Ethernet) NEM modules. The x6450 blades in the chassis have Solaris 10 u7 installed. In our tests, we used two blades in the following manner: each blade had only one network interface connected (and plumbed in the OS level) and each blade was connected to a different NEM. Both blades were connected to the same gigabit switch. There were two other test systems (x2270-s) also connected to the same switch. We ran the following tests:
both the application client and server are on the same blade (i.e. application is using the loopback network interface): n operations/second (the number here is irrelevant, only the relation to other test numbers are relevant)
application client on first blade, server on second blade: 0.77 x n operations / second (i.e. 23% degradation)
application client on blade, server on x2270: 0.77 x n operations /second (i.e. 23% degradation)
application client on x2270, server on blade: 0.75 x n operations /second (i.e. 25% degradation)
We ran similar tests with the x2270. We did not see any difference between running the test on the loopback interface vs. running the test between two x2270-s.
Is there a known issue with these network modules or their drivers?
Edited by: kbertold on Jul 14, 2009 11:54 AM

Just use a high interval and you see how much throughput you have in that time
example: dlstat vnet1 120
Best regards,
Marcel

Mac network performance issue

I've just had a new ISP hookup. My WiFi is blazing fast. But my Mac network speed is slowwwww.
Speedtest.net just reported:
Ping: 253ms
Download: 1.12Mbps
Upload: 0.43 Mbps
On my iPad via Wifi, I get:
Ping: 19ms
Download: 13.88Mbps
Upload: 0.90Mbps
My Mac I think dates back to about 2005/6, even though you would think 2008 based on sites like everymac.com. But I bought it long before 2008.
The only modification I've made to it has been to bump it from 2GB to 4GB memory. That slowed down the bootup, but not ridiculously.
How can I diagnose fix this network issue?

Please read this whole message before doing anything.
This procedure is a diagnostic test. It’s unlikely to solve your problem. Don’t be disappointed when you find that nothing has changed after you complete it.
The purpose of the test is to determine whether the problem is caused by third-party software that loads automatically at startup or login, by a peripheral device, or by corruption of certain system caches.
Disconnect all wired peripherals except those needed for the test, and remove all aftermarket expansion cards. Boot in safe mode and log in to the account with the problem. Note: If FileVault is enabled on some models, or if a firmware password is set, or if the boot volume is a software RAID, you can’t do this. Ask for further instructions.
Safe mode is much slower to boot and run than normal, and some things won’t work at all, including sound output and Wi-Fi on certain iMacs. The next normal boot may also be somewhat slow.
The login screen appears even if you usually log in automatically. You must know your login password in order to log in. If you’ve forgotten the password, you will need to reset it before you begin. Test while in safe mode. Same problem? After testing, reboot as usual (i.e., not in safe mode) and verify that you still have the problem. Post the results of the test.

Oracle CPU Jan 2009 cause performance issue

I did installed Oracle CPU Jan 2009 on HP-UX machine. But once the installation is completed, users complaint saying that it takes more than 1 minutes to open a new ticket on that application. With this, it cause a backlog processess increase tremendously.
Will the CPU Jan 2009 patch cause any of the network performance issue:
Server: HP-UX Itanium 64 bit
Database: Oracle 10.2.0.3.0
Instances: 2 instances running on this server.
Edited by: user3858134 on Oct 26, 2009 9:30 PM

I believe the latest CPU Patch for Oracle 10.2.0.3 on HP is CPU Jan 2009 only.Don't you think your database should be on 10.2.0.4?
Anyways, do you have any benchmark statspack/AWR report? Can you compare that with the latest one? Do you see any difference?
Regards,
S.K.

AT&T USBConnect Lightning performance issues with Snow Leopard

I recently purchased the AT&T USBConnect Lightning wireless data adapter and attempted to install it on my 15" MBP but ran into global network performance issues afterward so I wanted to check with anyone here who may have the same adapter. I noticed that the adapter has software drivers preloaded on the device that need to be installed on my Mac. Looking at the notes, it refers only to Mac OSX 10.4.11 or better and no mention of Snow Leopard so I'm now very concerned. The notes state that once the drivers are installed, that I'm required to reboot my computer which leads me to think the drivers may have affected MTU and other settings that might also impact performance with my other network adapters.
Appreciate any feedback you might have with this scenario.
Thanks so much!

I give up

JCaps 5.1.3 Sun Solaris CPU performance issue

Folks,
We are experiencing a serious CPU performance issue on our Solaris server with HL7 projects deployed.
The projects consist of the sample HL7 inbound and outbound projects with an additional service sending to a batch local file external for writing journals.
The performance issue occurs when there is volume of data in the queues/topics. As we continue to deploy additional HL7 projects (usually about 6 interfaces), the CPU increases until it reached 100%.
This sanapshot is prstat when no date is transmitting through the interfaces (One inbound - one outbound):
B PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP
15598 jre 379M 177M sleep 59 0 2:49:11 3.1% eManager/74
21549 phs 1174M 1037M sleep 59 0 14:49:00 2.5% is_dm_phs/113
23090 phs 3456K 3136K cpu1 59 0 0:00:01 0.4% prstat/1
23102 phs 3792K 3496K sleep 59 0 0:00:00 0.2% prstat/1
21550 phs 46M 35M sleep 59 0 0:13:27 0.1% stcms.exe/3
1272 noaccess 209M 95M sleep 59 0 0:26:30 0.1% java/25
11733 jre 420M 212M sleep 59 0 1:35:40 0.1% java/34
131 root 4368K 2480K sleep 59 0 0:02:10 0.1% nscd/30
23094 phs 3064K 2168K sleep 59 0 0:00:00 0.1% bash/1
This sanapshot is prstat when data is transmitting through the interfaces(One inbound - one outbound):
PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP
21549 phs 1174M 1037M cpu1 20 0 14:51:20 88% is_dm_phs/113
15598 jre 379M 181M sleep 59 0 2:49:18 1.3% eManager/74
21550 phs 46M 35M sleep 49 0 0:13:29 1.2% stcms.exe/3
23090 phs 3456K 3128K cpu3 49 0 0:00:03 0.4% prstat/1
1272 noaccess 209M 95M sleep 59 0 0:26:30 0.1% java/25
11733 jre 420M 212M sleep 59 0 1:35:40 0.1% java/34
21546 phs 118M 904K sleep 59 0 0:01:21 0.1% isprocmgr_dm_ph/13
This sanapshot is prstat -L when data is transmitting through the interfaces (One inbound - one outbound):
PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/LWPID
21549 phs 1174M 1037M cpu1 41 0 0:00:45 22% is_dm_phs/13971
21549 phs 1174M 1037M sleep 51 0 3:31:06 21% is_dm_phs/1394
21549 phs 1174M 1037M run 51 0 3:14:16 20% is_dm_phs/1296
21549 phs 1174M 1037M sleep 52 0 3:14:13 19% is_dm_phs/1380
15598 jre 379M 181M sleep 50 0 1:49:57 3.1% eManager/4
21549 phs 1174M 1037M sleep 59 0 0:15:36 1.7% is_dm_phs/4
21550 phs 46M 35M sleep 59 0 0:10:52 1.0% stcms.exe/1
21549 phs 1174M 1037M sleep 59 0 0:10:45 0.9% is_dm_phs/6
15598 jre 379M 181M sleep 54 0 0:33:35 0.3% eManager/35
21549 phs 1174M 1037M sleep 59 0 0:03:34 0.3% is_dm_phs/5
21550 phs 46M 35M sleep 59 0 0:02:37 0.2% stcms.exe/2
21549 phs 1174M 1037M sleep 59 0 0:02:17 0.2% is_dm_phs/3
21549 phs 1174M 1037M sleep 59 0 0:02:17 0.2% is_dm_phs/2
Solaris 10 server details:
CPU's (4x900 Sparc III+)
4096 MB RAM
SunOS testican 5.9 Generic_118558-39 sun4u sparc SUNW,Sun-Fire-880
Disk: 6 internal Fujitsu 72GBs
swapspace on the server:
total: 4305272k bytes allocated + 349048k reserved = 4654320k used, 10190536k available
My sysadmin has run statistics (iostat, vmstat, psig, pmap, pfind, pstack, mpstat, etc.) - and has reported that the server is performing fine - with the exception of the CPU. It also looked like the swap space was not being utilized.
We have increased the MaxPerm value to 512, and increased the heapsize on isprocmgr_dm_phs to -Xmx2048m, and increased the heapsize on the domain to 2048 per KB 103824
We have also added the -d64 value (specific to Solaris) per the Deployment Guide.
We increased the value of Maximum Pool size in the JMS clients to 128 - per the deployment Guide.
We increased the swapspace on the server to 10Gb:
total: 4305272k bytes allocated + 349048k reserved = 4654320k used, 10190536k available
We have modified the tcpip and kernal parameters per the Sun Administration server 8.2 performance tuning guide:
core file size (blocks, -c) unlimited
data seg size (kbytes, -d) unlimited
file size (blocks, -f) unlimited
open files (-n) 8192
pipe size (512 bytes, -p) 10
stack size (kbytes, -s) 8192
cpu time (seconds, -t) unlimited
max user processes (-u) 29995
virtual memory (kbytes, -v) unlimited
None of these modificatons appear to increase performance.
Any help is appreciated.
Thanks
Rich...

Hi,
I noticed this behavior with the Alert + SNMP Agents installed but not configured. In this situation, the SNMP agent generates traps for all events, leading to high CPU using, even when nothing was processed. Are you in a similar case?
Regards

Sun JVM Performance Issue in Sun Solaris 10 (SPARC)

Hi,
Issue : Performance issue after the migration of a Java application from IBM-AIX 5 to Sun Solaris 10 (SPARC)
I am facing performance issue after the migration of a Java application from IBM-AIX 5.3 to Sun Solaris 10 (SPARC).
Normally the application takes less than 1 hour to complete the process in AIX, but after migration in Solaris the application is taking 4+ hours.
The Java version of IBM AIX is ,
java version "1.5.0"
Java(TM) 2 Runtime Environment, Standard Edition (build pap32dev-20051104)
IBM J9 VM (build 2.3, J2RE 1.5.0 IBM J9 2.3 AIX ppc-32 j9vmap3223-20051103 (JIT enabled)
The Java version of Solaris 10 is,
Java(TM) Platform, Standard Edition for Business (build 1.5.0_17-b04)
Java HotSpot(TM) Server VM (build 1.5.0_17-b04, mixed mode)
Description of Application
The application merges 2 XML files of size 300 MB each using DOM Parser and generates flat file according to certain business logic.No remote files are using for the file generation. There are two folders and around 200 XML file in each folders of similar names. The application loads 2 similar XML file at a time from each folder and Processes. Same way, the application processes all the 200 XML file pairs using loop.
The JVM Parameters are given below.
/usr/java5/bin/java -cp $CLASSPATH -Xms3072m -Xmx3072M com.db.mcc.creditderiv.GCDXMLTransProc
Here the extended swap memory in AIX is 3072 (3GB). After copying the same tode to Solaris, the
application started throwing java.lang.OutofMemoryError. So that we have increased the swap memory up to 12 GB.
Since 32bit Java allows maximum 4 GB extended memory we started using 64 Bit Java in Solaris using -d64 argument.
The Current JVM Parameter in Solaris is given below.
java -d64 -cp $CLASSPATH -Xms8192m -Xmx12288m com.db.mcc.creditderiv.GCDXMLTransProc ( 64 GB Swap Memory is available in the System)
We have tried the following options
1.       Extended heap size up to 12 GB using -xms and -xmx parameters and tried multiple -XX options. Earlier the application was working fine in AIX with 3.5 GB extended heap size. ( 64 GB Swap Memory is available in the System)
2.       Downloaded and installed the Solaris SPARC Patches from the website,
     http://java.sun.com/javase/downloads/index_jdk5.jsp
4. Downloaded and installed XML and XSLT patch from sun website
5.       Tried to run the Java in server mode using -server option.

A 64 bit VM is not necessarily faster than a 32 bit one. I remember at least on suggestion that it could be slower.
Make sure you use the -server option.
As a guess IBM isn't necessarily a slouch when it comes to Java. It might simply be that their VM was faster. Could have used a different dom library as well.
Could be an environment problem of course.
Profiling the application and the machine as well might provide information.

Performance issues (Oracle 9i Solaris 9)

Hi Guys,
How do I tell if my database is performing at its optimum level. We seem to be having perfomance issues on one of our applications. There are saying it's the database, network, etc.
Thank you.

Hi,
In order to determine whether or not your Database is having performance Issues,you will need to install and execute Statspack. Statspack is utility which provides information about the Performance Parameters of Oracle Database.
If you are already using statspack report for performance analysis post the snapshot of the report.........
Regards,
Prosenjit Mukherjee.

Performance issue of Jboss,JDK on solaris

Hi ,
We recently upgraded to solaris 10 (5.10) on sun sparc 64 bit machine along with jboss and jdk .
With solaris 10 , new jboss and new jdk our application performance has come down.
Old configuration
sun sparc 64 bit
solaris 9 (no patches applied )
oarcle 10.2.0.1 (no patches applied )
jboss 3.2.7
jdk 1.4
New configuration
sun sparc 64 bit
solaris 10 (no patches applied )
oarcle 10.2.0.1 (latest patches applied i.e. 10.2.0.4 )
jboss 4.2.0
jdk 1.5_12
we would like to know about any known Solaris10 , Oracle10g issues / patches to address performance issues.
and also would like to know about any specific tunable jboss ,jdk settings/parameters on solaris10 for improving the performance .
thanks a lot in advance.

I'm interested in the same question since we are seeing a similar situation.

Performance Issue: Central Instance not utilizing full network resources.

Hi All,
I have experienced a problem regarding network usage by central instance.
At random times, even at peak load, SAP application server utilizing not more than 0.30%-0.35% of network resources, resulting in very poor performance.
As Primary observation, i have checked at operating system level that no other service/application is consuming network at that time.
Please help, how i can monitor/check and enhance the network resources utilization to the fullest for better performance.
Thanks in advance.
Regards
Surjit

SAP application server utilizing not more than 0.30%-0.35% of network resources, resulting in very poor performance.
To be honest with you I can't see the network to been the performance issue unless you have serius trouble with network latency or maybe your system been located in a remote location.
You can check the average GUI time (in milliseconds) in ST03N, then you can make conclutions on if the info travelling to the frontend is been compromised because of network issue.
Also, usually comunication in between the instances should be based on a private fast VLAN to allow free traffic in between them
Regards
Juan

Performance issue studio 10 fortran compiler on 2500 blade solaris 10

Dear All,
We facing performance issue on sun blade 2500 with fortran (studio 10).
when we run our code it took 14 min. where same code tokk 1min IRIS/Apple Power Mac/HP/Linux .
if u any solution on this pl. mail me [email protected]
regards
Narayan

You'll need to provide more details before we can help you. You could start by using the performance analyzer to find the slowest parts of your program. You might also be using inappropriate flags (e.g., compiling without optimization would be a problem).

11.2.0.0 upgrade- performance issue when commits

We have upgraded our data warehouse system from 10.2.0.4 for 11.2.0.2. After upgrade ETL system has performance issue. Extraction is fine because we compared the explain plans before and after upgrade. However, loading is taking very long. In loading there is a commit after every 1000 records, that why its slow. Even when we import data with commit=y option its very slow. Is there any know issue on this?
Any help pl?
Thanks

We are solaris 10. ETL job on 10.2.0.3 was taking 2 hours and now its taking 11 hours.
I beleive commit is a issue because
1. Source query (Extraction) has same explain plan.
2. Source database is still on 10.2.0.3
3. Servers and network is same for 10g and 11g.
4. I already experienced COMMIT=y performance issue in import.
Thanks

Performance issues with LOV bindings in 3-tier BC4J architecture

We are running BC4J and JClient (Jdeveloper 9.0.3.4/9iAS 9.0.2) in a 3-tier architecture, and have problems with the performance.
One of our problems are comboboxes with LOV bindings. The view objects that provides data for the LOV bindings contains simple queries from tables with only 4-10 rows, and there are no view links or entity objects to these views.
To create the LOV binding and to set the model for the combobox takes about 1 second for each combobox.
We have tried most of tips in http://otn.oracle.com/products/jdev/tips/muench/jclientperf/index.html, but they do not seem to help on our problem.
The performance is OK (if not great) when the same code is running as 2-tier.
Does anyone have any good suggestions?

I can recommend that you look at the following two bugs in Metalink: Bug 2640945 and Bug 3621502
They are related to the disabling of the TCP socket-level acknowledgement which slows down remote communications for EJB components using ORMI (the protocol used by Oracle OC4J) to communicate between remote EJB client and server.
A BC4J Application Module deployed as an EJB suffers this same network latency penalty due to the TCP acknowledgement.
A customer sent me information (that you'll see there as a part of Bug# 3621502) like this on a related issue:
We found our application runs very slow in 3-Tier mode (JClient, BC4J deployed
as EJB Session Bean on 9iAS server 9.0.2 enterprise edition). We spent a lot
of time to tune up our codes but that helped very little. Eventually, we found
the problem seemed to happen on TCP level. There is a 200ms delay in TCP
level. After we read some documents about Nagle Algorithm, we disabled a
registry key (TcpDelAckTicks) in windows2000 on both client and server. This
makes our program a lot faster.
Anyway, we think we should provide our clients a better solution other than
changing windows registry for them, for example, there may be a way to disable
that Nagle's algorithm through java.net.Socket.setTcpNoDelay(true), in BC4J,
or anywhere in our codes. We have not figured out yet.
Bug 2640945 was fixed in Oracle Application Server 10g (v9.0.4) and it now disables this TCP Acknowledgement on the server side in that release. In the BugDB, I see backport patches available for earlier 9.0.3 and 9.0.2 releases of IAS as well.
Bug 3621502 is requesting that that same disabling also be performed on the client side by the ORMI code. I have received a test patch from development to try out, but haven't had the chance yet.
The customer's workaround in the interim was to disable this TCP Acknowledgement at the OS level by modifying a Windows registry setting as noted above.
See Also http://support.microsoft.com/default.aspx?kbid=328890
"New registry entry for controlling the TCP Acknowledgment (ACK) behavior in Windows XP and in Windows Server 2003" which documents that the registry entry to change disable this acknowledgement has a different name in Windows XP and Windows 2003.
Hope this info helps. It would be useful to hear back from you on whether this helps your performance issue.

Performance Issues with Acrobat Reader 11.0.0.2 when secure mode is enabled

Hello All,
We are experiencing sporadic issues with Acrobat 11.0.0.2 across our domain, users are reporting performance issues when opening PDF documents whether locally or from a network share.
We have found that turning off Secure Mode helps towards reducing this delay and in the cases it doesn't we are repairing the installation and/or reinstalling the application.
Due to the security implications we need to leave this turned on, I am wondering if anyone has encountered this issue and what steps were taken towards resolving it?
I also wonder whether the white list function in the new release 11.0.0.3 would be a solution to this issue?
Kind Regards,
Ryan McCarty

No probelm, so....
We had no problems with Adobe Reader 9 and 10, we encountered the issues when upgrading to 11.0.0.2.
Initially we found that turning off the Protected Mode, helped but did not resolve the issue.
We tried;
1. Turn off protected mode - issue still present
2. Clearing the recent file registry using the below registry path and deleting the keys underneath it.
HKEY_CURRENT_USER\Software\Adobe\Acrobat Reader\11.0\AVGeneral\cRecentFiles (this does not turn recent files off permanently). - works but needs clearing regularly
3. Turning off welcome screen by creating - HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Adobe\Acrobat Reader\11.0\FeatureLockDown\cWelcomeScreen - works to improve app open speed.
4. uninstall/reinstall of 11.0.0.2 - works most likley due to the recent files being cleared.
5. upgrade to 11.0.0.3 - issue still present
Following reboots the issue is still present.
When Adobe Reader is the only application open this issue is still present.
As mentioned I have no systems available which I could test this issue using 11.0.0.1 as we have fixed them, albeit temporarily using the reinstall method.
I am concious that this issue is going to reoccur once that cache (recent files) builds back up because the fix above (#2) is clearing the recent files cache NOT disabling it.

Performance issues with Homesharing?

I have a Time Capsule as the base station for my wireless network, then 2 Airport Express setup to extend the network around the house, an iMac i7 as the main iTunes Library and couple of iPads, and a couple of Apple TVs. Everything has the latest software, but I have several performance issues with Home sharing. I've done several tests making sure nothing is taking additional bandwidth, so here are the list of issues:
1) With nothing else running, trying playing a movie via home sharing in an iPad 2 which is located on my iMac, it stops and I have to keep pressing the play button over and over again. I typically see that the iPad tries to download part of the movie first and then starts playing so that it deals with the bandwidth, but in many cases it doesn't.
2) When trying to play any iTunes content (movies, music, photos, etc) from my Apple TV I can see my computer library, but when I go in on any of the menus, it says there's no content. I have to reboot the Apple TV and then problem fixed. I's just annoying that I have to reboot.
3) When watching a Netflix movie on my iPad and with Airplay I send the sound to some speakers via Airplay through an Airport Express. At time I lose the connection to the speakers.
I've complained about Wifi's instability, but here I tried to keep everything with Apples products to avoid any compatibility issues and stay within N wireless technology, which I understood it was much more stable.
Has anyone some suggestions?

Hi,
you should analyze the db after you have loaded the tables.
Do you use sequences to generate PKs? Do you have a lot of indexex and/or triggers on the tables?
If yes:
make sure your sequence caches (alter sequence s cache 10000)
Drop all unneeded indexes while loading and disable trigger if possible.
How big is your Redo Log Buffer? When loading a large amount of data it may be an option to enlarge this buffer.
Do you have more then one DBWR Process? Writing parallel can speed up things when a checkpoint is needed.
Is it possible using a direct load? Or do you already direct load?
Dim

Solaris 9, V210, bge0 network performance issue

Similar Messages

Maybe you are looking for