OS paging versus PRIV mode
Hi Experts,
We always found there are users using more than 2GB for transaction, so we did the below changes in our system to avoid fragmentation and paging of extended memory
1.Instead of allowing a process to grab 2 GB of (shared) extended memory before going in PRIV mode, the limit will be reduced to 800 MB.As a counterpart, private heap memory limits will be increased from 2 GB to 3,2 GB.
2.So total maximum memory per user is still 4 GB, but high memory users will have to "pay for it"
3. The advantage is that private memory is given back to the operating system once the transaction is finished, whereas shared extended memory is never given back (by design).
our assumption is we would never see MaxUse of extended memory > 12 GB in future.so users are allowed getting PRIV mode from time to time in our production.
But now we are getting some other problem.We are getting too much OS paging (according SE06)
Just what to confirm here with the experts
1. This PRIV mode of the WP uses the needed memory exclusively, the WP will not be interrupted and so uses much CPU, so Priv mode causes the high OS paging.?
2. Does SE06 includes the OS paging with IO?
3. What are the other options we do have to reduce this high paging
Please guide
Thanks,
Jansi
1. This PRIV mode of the WP uses the needed memory exclusively, the WP will not be interrupted and so uses much CPU, so Priv mode causes the high OS paging.?
PRIV mode or not, paging will occur if there is not enough memory. Even in PRIV mode, WP will be interrupted at O/S level if O/S follows round-robin method(UNIX) of assigning tasks to CPU.
2. Does SE06 includes the OS paging with IO? I assume yes.
3. What are the other options we do have to reduce this high paging
Here are options:
1) Tune ABAP programs.
2) Add more memory. How much (AVG)free/min/max memory you see every hour in st06? If you see really low memory almost every hour every day, then I would either try to add more memory or tune ABAP programs or review memory allocation to different components(DB/SAP system(s)/Any other application/Java if dual stack) more carefully and revise allocation if needed or perform all of the above.
Bala
Similar Messages
-
Hi, we have problem with proccesses going to PRIV mode (SAPHTTP), extended memory:
Roll, extended and heap memory EXTM
ztta/roll_area 3000000 Byte Roll area per workprocess (total)
ztta/roll_first 1 Byte First amount of roll area used in a dialog WP
ztta/short_area 1600000 Byte Short area per workprocess
rdisp/ROLL_SHM 16384 8 kB Part of roll file in shared memory
rdisp/PG_SHM 8192 8 kB Part of paging file in shared memory
rdisp/PG_LOCAL 150 8 kB Paging buffer per workprocess
em/initial_size_MB 4092 MB Initial size of extended memory
em/blocksize_KB 4096 kB Size of one extended memory block
em/address_space_MB 4092 MB Address space reserved for ext. mem. (NT only)
ztta/roll_extension 2000000000 Byte Max. extended mem. per session (external mode)
abap/heap_area_dia 2000000000 Byte Max. heap memory for dialog workprocesses
abap/heap_area_nondia 2000000000 Byte Max. heap memory for non-dialog workprocesses
abap/heap_area_total 2000000000 Byte Max. usable heap memory
abap/heaplimit 40000000 Byte Workprocess restart limit of heap memory
abap/use_paging 0 Paging for flat tables used (1) or not (0)
have You any advice?
ThanksHi,
A process going into PRIV means that either it has exhausted its quota of extended memory (ztta/roll_extension) or that the total extended memory (em/initial_size_MB) has been used up by other sessions.
The parameters look pretty standard, but you must determine the memory usage at the time you see a PRIV process. Do this via SM04 -> Goto -> Memory.
Another piece of advice: when you ask a question about memory-related issues, please always mention your operating system (Windows, UNIX, Linux, ...), O/S release and SAP release. SAP memory management has quite a few platform dependencies in it.
Regards,
Mark -
Stop DIA work process with PRIV mode in my program
Hello,
I write Program that display in the end an alv report and when i display it after same time my dia wp get in PRIV mode
I search about it and I didn't find the solution but it happend only with my program this reject the memory parameter that cause the problem
Thanks for your helpHi,
>I think this not the problem because in the dev and the qa system it won't get in priv thanks
>mode and the have the same value in the parameter ztta/roll_extension and the qa is the copy of the prod but 3 month old
And these 3 months difference can perfectly explain why you have the problemn on production and not on QA.
Private modes are generated usually because of badly designed programs (including standard ones...).
ALV is not designed to display a huge amoiunt of data.
Beware also of internal tables : Abap programmer usually abuse internal tables thinking memory is an unlmimited ressource...
Just a few ideas.
Regards,
Olivier -
Stop DIA work process with PRIV mode.
Hi,
How to stop DIA workprocess with PRIV mode. by searching notes, I find it is possible to setl the max time and max number of the private memory.
Kind regards,
MasaakiHello Masaaki,
Before deciding whether to wait for the process to finish by itself or not, i would suggest that you find out the program which is causing the work process to go into PRIV mode. Once this is done, as your technical guys (ABAP) to analyse why this is happening (is the program logic inefficient and hence using so much memory etc).
You have to note that (Dialog) work processes going into PRIV mode is not a healthy sign for the SAP application server.
This is beacuse once the WP goes into PRIV mode it is not available to any other transactions, and we only have a limited number of work processes.
Please make sure the parameters below are defined in RZ10 instance profile:
rdisp/max_priv_time - used to define the maximum time that a work process can remain in PRIV mode. (After this the work processes is terminated and restarted). This is reasonable because beyond the time limit set, it is imperative to assume that the program using the work process is in efficient. (in case a program really needs that much time, it needs to be scheduled as a background job).
rdisp/wppriv_max_no- this parameter defines the maximum number of work processes that can be in PRIV mode. Very useful.
Also, make sure that a parameter ABAP/heaplimit is defined. This ensures that if a program eats up a certain amount of heap memory, the work process being used can be terminated.
And finally please do check if your extended memory allocation is not big enough. the parameter em/initial_size_mb defines the fixed size of extended memory. Increasing this size only if you find that a lot of processes are frequently going into PRIV mode.
Hope this helps.
Regards,
Prashant -
Hi Gurus,
WP's are going to priv mode. user is not accepting to kill the process or reduce the number of records. what is the best way to do as a basis consultant?
Thanks in advance.Dear Sahad,
<PRIV> mode will be used if the work process starts to allocate heap - or private,
thus PRIV - memory. This is not necessarily a bad thing. There are two scenarios
that may lead to such a system state:
1. There is a small number of users that create huge contexts. In this case, it
will be OK to have such users allocate memory from heap and exclusively lock
a work process for them (this is a side-effect of the PRIV state, as heap memory
content can not be seen by another process, and the user context can no longer
be 'rolled' into another work process. The limit when this will happen is either given
by the ressource limit (em/initial_size_MB, EM/TOTAL_SIZE_MB) which is the
overall amount of extended memory (EM). EM is a shared memory. The other limit
is a user specific quota given by ztta/roll_extension (or ES/SHM_MAX_PRIV_SEGS
for AIX, when ES/TABLE=SHM_SEGS is used). As I said, this is normally OK, and
a small number of PRIV processes is not a problem at all.
2. The user contexts are equally big, but there is not enough EM available on the
instance. This will force EVERY user context into PRIV, as soon as the EM is
exhausted. The fact that EM is exhausted will be seen in transaction ST02 (look
for SAP Memory, Extended memory). Also check with transaction SM04, Goto,
Memory, the size of the user contexts. In this case, increasing EM, or reducing the
amount of users on the instance are your choices.
As I said, if this is only one or two work processes, you must not worry about them.
Best Regards,
Tim -
RE: Workprocess in PRIV mode
Hi all,
We are facing problem in SAP. The performance of the system is very low and all the dialog workprocess are entering 'PRIV' mode in 'ONHOLD' state.
Can you please say me what are the possible reasons for process to enter 'PRIV' mode?
Thanks in advance
Regards,
Sowmya KHi,
If all the work processes are going in the PRIV mode, it is very possible to be Kernel bug which you can check with SAP.
Understanding the HEAP memory better, If a dialog work process has used up the roll area assigned to it and the extended memory, private memory is assigned to the work process. The work process goes into PRIV mode (private).
In the parameter settings, ensure that not too many dialog work processes are simultaneously in the PRIV mode, since this lowers system performance.
The other Work Process types (background, update, enqueue and spool work processes) are assigned heap memory after the roll area is used up. Allocating Extended Memory is not useful since no context change occurs for these processes.
Other processes cannot use private (heap) memory. After releasing the assigned memory, the operating system still considers the (virtual) memory as being occupied by the allocating process. These characteristics of heap memory require that:
1. The work process can be run in PRIV mode (private) when the local memory is assigned. This means that the work process is reserved for processing the current user context until the context releases the work process again when the request has ended.
2. The work process, if it has used a lot of private memory, is restarted when the user context is terminated and the local memory is returned. The restart makes the local memory available again for other processes. A restart occurs if a work process uses more local memory than is defined in the parameter abap/heaplimit. The mechanism is displayed again there.
Automatic Termination of PRIV Mode Processes:
If too many dialog work processes run in the PRIV mode in an SAP application server, performance problems arise. The normal processing of user inquiries is slowed down by the PRIV mode work processes.
The SAP System offers a mechanism that lets you terminate non-active dialog processes in the PRIV mode. This mechanism helps reduce performance problems.
The mechanism works as follows:
In the PRIV mode, a maximum number (n) of dialog work processes can run without any time restrictions.
To determine this number n, set the value of the profile parameter rdisp/wppriv_max_no. If a value is not entered, the SAP System determines this number using the following formula:
the number (n) is set to the greater of the 2 following values:
Number of dialog work processes minus 5
1
If more than (n) dialog work processes are active and the time span set in parameter rdisp/max_priv_time (default 600 seconds) has elapsed, the transaction for that PRIV process, which has spent the longest possible time in PRIV mode, is reset.
Hope this information helps in sorting out the issue.
Thanks
Tanuj
(*Award point if useful.) -
Dialogue workprocesses going in PRIV mode
Dear all,
More than 95% of dialogue work processes are going in PRIV mode
The extended memory in ST02 is 100% full
Nobody is able to do nothing
Below pasted are the values from the instance profile
ztta/roll_extension 4000000000
rdisp/wppriv_max_no 1
zcsa/db_max_buftab 10000
sap/bufdir_entries 6000
rsdb/obj/large_object_size 16384
rsdb/obj/max_objects 4000
rsdb/obj/buffersize 8192
rsdb/ntab/irbdsize 9000
rsdb/ntab/ftabsize 45000
rsdb/ntab/entrycount 30000
rsdb/cua/buffersize 6000
abap/buffersize 600000
zcsa/presentation_buffer_area 6600000
zcsa/table_buffer_area 30000000
ztta/parameter_area 8000
auth/auth_number_in_userbuffer 9000
Any suggestions?Dear all,
Problem was with the following two parameters: --
em/initial_size_MB = 512
em/max_size_MB = 512
De-activating these 2 parameters in instance profile, solved the problem
Thank you all
Would like to add some more here just in case if anyone wants some more details: --
If a Dialog work process has used its quota of Extended Memory (EM) defined by parameter ztta/roll_extension OR all EM has been used up by all the other work processes (em/initial_size_MB), then the task starts to allocate HEAP up to quota (per work process) of abap/heap_area_dia.(this is PRIV mode).
To investigate, check out the dev_w* trace file, to check which program cause work process go into PRIV
Also, follow what Markus has suggested
You can try to set rdisp/wppriv_max_no to prevent this kind problem.
Exmaple: If you set it as 5, will limit only 5 work processes go to PRIVmode.
Please also update your kernel patch to latest version, if you haven't done so.
See also note: 79435 Automatic resetting from PRIV mode
Rgds -
Workprocess are stopped due to PRIV mode
hi experts,
one of applicatin server most of the workprocess have been stopped due to reason PRIV mode. I killed the session of some users but the problem persisting again.
please suggest
regards,
Sudhakar KPlease check the memory parameters like ,
abap/heap_area_dia
abap/heap_area_nondia
abap/heap_area_total
abap/heaplimit
and check the recommendations ( RZ11 ) and increase to 2000000000 (UNIX) if u have high memory requirements and more concurrent users doing heavy tasks.
These thresholds can be determined with
the test program 'memlimits'
cwrci:cwradm 5> memlimits
SAP R/3 address space configuration test tool V3.2 (98/01/14)
=========================================================================
Restart, set early swap allocation (PSALLOC=early)
SAP R/3 address space configuration test tool V3.2 (98/01/14)
=========================================================================
Check the maximum data size per process (malloc)
Check the available swap space (malloc in several processes)
Process 1974488 allocating ... Size = 4096MB Total: 4096MB
Upper size limit 4096 reached. Finish checking
Total available swap space = 4096MB
Check the maximum size of mapped file (mmap anonymous,dev/zero)
Check protection operations on this area (protect)
Trying to mmap 512MB ... successfull
Trying to mmap 768MB ... successfull
Trying to mmap 1024MB ... successfull
Trying to mmap 1280MB ... successfull
Trying to mmap 1536MB ... successfull
Trying to mmap 1792MB ... successfull
Trying to mmap 2048MB ... successfull
Trying to mmap 2304MB ... successfull
Trying to mmap 2560MB ... successfull
Trying to mmap 2816MB ... successfull
Trying to mmap 3072MB ... successfull
Trying to mmap 3328MB ... successfull
Trying to mmap 3584MB ... successfull
Trying to mmap 3840MB ... successfull
Trying to mmap 4096MB ... successfull
Upper size limit 4096 reached. Finish checking
Trying to mprotect 4096MB ... successfull
Maximum mapped file size: 4096MB
Maximum mprotect size: 4096MB
R/3 parameter em/initial_size_MB up to 4096 permitted
Check the maximum address space per process usable
both by process local memory and mapped file
Maximum address space ( mmap(4096 MB)+ malloc(1676MB) ): 5772MB
Result
Maximum heap size per process........: 4096 MB
measurement limited to 4096 MB
this value is probably limited by swap space
Maximum mapped file size (mmap)......: 4096 MB
measurement limited to 4096 MB
Maximum protectable size (mprotect)..: 4096 MB
measurement limited to 4096 MB
em/initial_size_MB > 4096 MB will not work
Maximum address space per process....: 5772 MB
Total available swap space...........: 4096 MB
measurement limited to 4096 MB
main memory size x 3 recommended , minimum 1 GB
cwrci:cwradm 6> -
SSID on FlexConnect versus Local mode APs???
Hello!
A collegue of mine and I discussed the different ways we could deliver a SSID on a customers APs on their geographically different sites ......
The customer have a WLC5508 (r7.6) and (mostly) AP1142.
All of the APs are in FlexConnect mode
Two SSIDs are centrally switched
One SSID are FlexConnect on all the sites with a local VLAN
Now we would like to deploy a new SSID which should be centrally switched on all the sites, except for one site ... So the problem is that the SSID need support for FlexConnect for one site but should be centrally switched on all the other sites. And on these sites the APs are also in FlexConnect mode...
Is there a way to do this??? We have been looking around the settings for WLAN, APs, FlexConnect groups etc and cannot figure this out! :-)
Best Regards
Göran Blomqvist
TDC
SwedenHow about creating two SSID profiles for the same SSID Name. One with WLAN-ID > 16 & configured it for FlexConnect local switching. Then create an AP group for the particular branch & map that SSID (the one with local switching) to that.
For other SSID (without local switching) you can map to all other branch AP (if you have specific group). If you have ap in default-apgroup then as long as you choose WLAN ID < 16, it should be available in all other branches
Give it a try & see.
HTH
Rasika
**** Pls rate all useful responses **** -
Hi All - Thanks for the excellent info I've gotten off these
forums. I am a newbie and have a basic question. Is it better to
use the Layout Mode or the Table Mode (Standard) for building Web
pages? And what are some of the positives for each and/or the
negatives?
I have had very good luck using Layers—easier to move
around and size, but a few people I've spoken with say to stay with
the Tables.
Thanks,
ErnieNice of you to say so, John. Similarly, it was the many
generous people on
this very forum who 'raised me' as well....
Murray --- ICQ 71997575
Adobe Community Expert
(If you *MUST* email me, don't LAUGH when you do so!)
==================
http://www.dreamweavermx-templates.com
- Template Triage!
http://www.projectseven.com/go
- DW FAQs, Tutorials & Resources
http://www.dwfaq.com - DW FAQs,
Tutorials & Resources
http://www.macromedia.com/support/search/
- Macromedia (MM) Technotes
==================
"Tarvardian" <[email protected]> wrote in
message
news:[email protected]...
> Well, two minutes late anyway. But it's proof that many
of us out here
> take your advice and lessons seriously. I know I've
learned much just
> from reading your posts.
>
> John
>
>
> "Murray *ACE*" <[email protected]>
wrote in message
> news:[email protected]...
>> Hehe.... A day late and a dollar short!
>>
>> --
>> Murray --- ICQ 71997575
>> Adobe Community Expert
>> (If you *MUST* email me, don't LAUGH when you do
so!)
>> ==================
>>
http://www.dreamweavermx-templates.com
- Template Triage!
>>
http://www.projectseven.com/go
- DW FAQs, Tutorials & Resources
>>
http://www.dwfaq.com - DW FAQs,
Tutorials & Resources
>>
http://www.macromedia.com/support/search/
- Macromedia (MM) Technotes
>> ==================
>>
>>
>> "Murray *ACE*"
<[email protected]> wrote in message
>> news:[email protected]...
>>> In my opinion, there are three serious problems
with Layout Mode -
>>>
>>> 1. Perhaps most importantly, it sits between you
and *real* HTML
>>> tables,
>>> and fools you into believing that concepts like
"layout cell" and
>>> "autostretch" really mean something. They do
not. As long as you use
>>> Layout Mode, you'll never learn one of the most
important things for new
>>> web
>>> developers - how to build solid and reliable
tables.
>>>
>>> 2. Actually, #1 wouldn't be *so* bad, except
that the code that is
>>> written
>>> by Layout Mode is really poor code. For example,
a layout table
>>> contains
>>> MANY empty rows of cells. This can contribute to
a table's instability.
>>> In addition, if your initial positioning of the
table's cells is a bit
>>> complex,
>>> Layout Mode will throw in col- and rowspans
aplenty as it merges and
>>> splits
>>> cells willy-nillly to achieve the pixel-perfect
layout you have
>>> specified. Again,
>>> this is an extremely poor method for building
stable tables, because it
>>> allows
>>> changes in one tiny cell's shape (i.e,
dimensions) to ripple through the
>>> rest
>>> of the table, usually with unexpected and
sometimes disastrous
>>> consequences.
>>> This is one of the primary reasons for the final
result's fragility -
>>> read this -
>>>
>>>
http://apptools.com/rants/spans.php
>>>
>>> 3. The UI for Layout Mode is beyond confusing -
many options that you
>>> might
>>> want to use are inaccessible, e.g., inserting
another table, or layer
>>> onto the page.
>>>
>>> I can understand the new user's desire to use
this tool to make their
>>> life easier,
>>> but the cost is just too heavy in my opinion.
>>>
>>> To make good tables, keep it simple. Put a table
on the page, and begin
>>> to
>>> load your content. If you would want a different
table layout, instead
>>> of
>>> merging or splitting cells, consider stacking
tables or nesting simple
>>> tables instead, respectively.
>>>
>>> And above all, do not try to build the whole
page with a single table!
>>>
>>> To read more about this approach, visit the DW
FAQ link in my sig, and
>>> run
>>> through the table tutorials.
>>>
>>>> I have had very good luck using
Layers?easier to move around and size,
>>>> but a
>>>> few people I've spoken with say to stay with
the Tables.
>>>
>>> Using Layers as a primary layout methodology is
a slippery slope to
>>> catastrophe. Here's why -
>>>
>>>
http://www.great-web-sights.com/g_layer-overlap.asp
>>>
>>>
>>> --
>>> Murray --- ICQ 71997575
>>> Adobe Community Expert
>>> (If you *MUST* email me, don't LAUGH when you do
so!)
>>> ==================
>>>
http://www.dreamweavermx-templates.com
- Template Triage!
>>>
http://www.projectseven.com/go
- DW FAQs, Tutorials & Resources
>>>
http://www.dwfaq.com - DW FAQs,
Tutorials & Resources
>>>
http://www.macromedia.com/support/search/
- Macromedia (MM) Technotes
>>> ==================
>>>
>>>
>>> "DreemWeeve"
<[email protected]> wrote in message
>>> news:[email protected]...
>>>> Hi All - Thanks for the excellent info I've
gotten off these forums. I
>>>> am a
>>>> newbie and have a basic question. Is it
better to use the Layout Mode
>>>> or the
>>>> Table Mode (Standard) for building Web
pages? And what are some of the
>>>> positives for each and/or the negatives?
>>>>
>>>> I have had very good luck using
Layers?easier to move around and size,
>>>> but a
>>>> few people I've spoken with say to stay with
the Tables.
>>>>
>>>> Thanks,
>>>>
>>>> Ernie
>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>
> -
Problem with media print styles in debug versus nondebug mode
Hi,
I am trying to debug why the same page (and code) will print one way on one server and another way on another server. These servers are a dev and test server with the same version of cf 9 installed.
I am working on a page that has two divs floated left next to each other.
[div 1] [div 2]
We had problems printing this setup in IE 8, so I created a CSS file for media type print that removes the float from div2.
So in my page I have:
<link rel="stylesheet" type="text/css" href="../common/main.css" media="screen"/>
<link rel="stylesheet" type="text/css" href="../common/mainPrint.css" media="print"/>
In main.css the class is:
.documentArea {
margin: 2.5em 1em 0 1em;
float:left;
In mainPrint.css the class is:
.documentArea {
margin: 2.5em 1em 0 1em;
On the Dev server, I am able to view the page correctly in IE and it prints correctly with div 2 below div 1:
[div 1]
[div 2]
When I move the SAME code to the our Test server and view it in the same IE browser, the display of the page is correct but the printed page only displays div 1 and not div 2 at all.
After some investigations, I discovered that the the print CSS worked when debugging was on and not when it was off. We turned on debugging on the Test server and the CSS worked correctly for printing. I also changed the name of the style to my name (leslie) and it still didn't work. I thought perhaps a default style in CF might be overriding it.
I am wondering if there are default CSS files in ColdFusion.
If you have ANY suggestions as to what is happening I would greatly appreciate it!
Regards,
LeslieJust to clarify ....
If the ERROR_CODE parameter is made an 'OUT' parameter instead of an 'IN OUT' parameter the correct return value is given for the OracleCommand invocation but the WRONG value is still returned for the EF Model invocation i.e. just changing the parameter from 'IN OUT' to just 'OUT' does not fix the problem. -
Extended memory 50% free,but work process reach PRIV
hi,
many dialog work process reach PRIV mode though extended memory is 50% free.
profile parameters as below :
ztta/roll_first 1 MB
ztta/roll_extension 2GB
ztta/roll_area 6MB
abap/heap_area_dia 2GB
can anyone throw light ?Hi Krishna,
In addition to the parameter you have mentioned check these parameter as well in the system.
A process going into PRIV means that either it has exhausted its quota of extended memory (ztta/roll_extension) or that the total extended memory (em/initial_size_MB) has been used up by other sessions.So increase these 2 parameter to start with and then tune others.
rdisp/ROLL_SHM 32768 8 kB Part of roll file in shared memory
rdisp/PG_SHM 16384 8 kB Part of paging file in shared memory
rdisp/PG_LOCAL 150 8 kB Paging buffer per workprocess
em/initial_size_MB 61440 MB Initial size of extended memory
em/blocksize_KB 4096 kB Size of one extended memory block
em/address_space_MB 4096 MB Address space reserved for ext. mem. (NT only)
ztta/roll_extension 2000000000 Byte Max. extended mem. per session (external mode)
abap/heap_area_total 64424500000 Byte Max. usable heap memory
abap/heaplimit 40000000 Byte Workprocess restart limit of heap memory
Also,consider setting rdisp/wppriv_max_no and rdisp/max_priv_time,so that system performance is not much impacted.
http://help.sap.com/saphelp_nw04/helpdata/en/7a/caa6f3bfdb11d188b30000e83539c3/content.htm
Regards,
Ashutosh -
Does back ground jobs work process go to private mode or not
Hi Experts,
Can any one tell me
Does back ground jobs work process go to private mode or not
Thanks and Regards
Dan !!Hi Dan,
I do not think background work processes would go into PRIV mode.
the orginal information was remove by Admin because provided by Cut and Paste from SAP Online Help.
you will find it <a href="http://help.sap.com/saphelp_47x200/helpdata/en/7a/caa6f3bfdb11d188b30000e83539c3/frameset.htm">here</a>
The distribution, participation in distributing or otherwise sending of this material is against the law. The material you are requesting is copyrighted material and available ONLY to customers of SAP. If you need such documents from the SAP Service Marketplace then you must have "S" user ID and login yourself to retrieve this material. If you do not have a "S" user ID then you should contact your own internal company groups responsible and request one or request them to retrieve the document for you.
By participating and sending such documents you are at risk for legal action and a removal of your account here on SDN and BPX.
DO NOT send material via email such as this! Further actions will result in officials within your company and your SAP Sales Account being notified and could result in legal action against you as an individual. -
DpVmCheck: max number of priv VMs exceeded (1 0, timeout)
Hello,
In our CRM system EHP1 FOR CRM 7 ABAP ,users get this screen mainly for big contracts at different stages like collision run,saving.
Status before error started
I have already gone through notes 1757252
But the parameter rdisp/max_vm_priv_time and rdisp/max_vm_priv_no doesn't exist in our system. Instead is there any other parameter which we can use in our release ?
I could find these errors when the timeout screen occurred :
DEV_DISP
Wed Jul 23 15:26:08 2014
DpVmCheck: max number of priv VMs exceeded (1 > 0, timeout), reset T59/M0/U13880/V4
Wed Jul 23 15:26:28 2014
DpVmCheck: max number of priv VMs exceeded (1 > 0, timeout), reset T59/M0/U13880/V4
Wed Jul 23 15:26:46 2014
*** ERROR => DpRqCheck: T59 in state TM_SLOT_FREE [dpxxdisp.c 8006]
***LOG Q0G=> DpRqBadHandle, bad_req ( DIA) [dpxxdisp.c 6125]
*** ERROR => DpRqBadHandle: BAD REQUEST - Reason: DpRqCheck failed (line 7037): [dpxxdisp.c 6127]
-IN-- sender_id DISPATCHER tid 59 wp_ca_blk -1 wp_id -1
-IN-- action SEND_TO_WP uid 13880 appc_ca_blk -1 type DIA
-IN-- new_stat NO_CHANGE mode 0 len 0 rq_id 24366
-IN-- req_info DP_CANCEL_PRIV
-IN-- msg_name -
*** ERROR => DpRqCheck: T59 in state TM_SLOT_FREE [dpxxdisp.c 8006]
***LOG Q0G=> DpRqBadHandle, bad_req ( DIA) [dpxxdisp.c 6125]
*** ERROR => DpRqBadHandle: BAD REQUEST - Reason: DpRqCheck failed (line 7037): [dpxxdisp.c 6127]
-IN-- sender_id DISPATCHER tid 59 wp_ca_blk -1 wp_id -1
-IN-- action SEND_TO_WP uid 13880 appc_ca_blk -1 type DIA
-IN-- new_stat NO_CHANGE mode 0 len 0 rq_id 24502
-IN-- req_info DP_CANCEL_PRIV
-IN-- msg_name -
*** ERROR => DpRqCheck: T59 in state TM_SLOT_FREE [dpxxdisp.c 8006]
***LOG Q0G=> DpRqBadHandle, bad_req ( DIA) [dpxxdisp.c 6125]
*** ERROR => DpRqBadHandle: BAD REQUEST - Reason: DpRqCheck failed (line 7037): [dpxxdisp.c 6127]
-IN-- sender_id DISPATCHER tid 59 wp_ca_blk -1 wp_id -1
-IN-- action SEND_TO_WP uid 13880 appc_ca_blk -1 type DIA
-IN-- new_stat NO_CHANGE mode 0 len 0 rq_id 24691
-IN-- req_info DP_CANCEL_PRIV
-IN-- msg_name -
sm21:
Q0 G Request (type DIA) cannot be processed
work process trace
please check the image
Thanks
ShradhaHello Divyanshu
It didnt create dev_vm.trc and VM_STACK_TRACE
And I am unable to provide dev_disp and dev_w* here due to size of the trace
But could the error be due to any of these parameters :
Parameters
Current
Changed
rdisp/max_vm_priv_time
2400
SERVLET=2400, RFC=2400
isp/max_vm_priv_no
3
SERVLET=3, RFC=3
rdisp/max_priv_time
600
3600
Maximum hold time for work processes in PRIV mode
rdisp/max_debug_lazy_time
600
3600
Maximum hold time for work processes in DEBUG mode
rdisp/appc_ca_blk_no
1000
1058
Buffer Size for CPI-C Communications
rdisp/wp_ca_blk_no
300
1058
Work Process Communication Blocks
vmcj/option/ps
960M
1460M
poolsize for the VMC
vmcj/option/gc/sharedWaterMark
90
70
VMC: watermark for shared garbage collector
vmcj/max_vm_heap_MB
64
128
max. vm heap
rdisp/keepalive
1200
3600
Timer for check of unused connections
rdisp/keepalive_timeout
60
150
Timeout for connection to SAPGUI
rdisp/wppriv_max_no
DEFAULT
16
Maximum number of work processes in PRIV mode -
Hi,
In our BI system, we want to avoid that dialog workprocess entering in private mode.
Now we have the next situations:
Number of dialog workprocess: 30
rdisp/wppriv_max_no: 25
rdisp/max_priv_time: 600
We had thought to change this parameters to:
rdisp/wppriv_max_no: 15
rdisp/max_priv_time: 3600
Do you think these new parameters are correct or that we can to have some problems whit them?.
Best RegardsHello Maximino,
as so often: It depends.
And as I wrote: You can't really compare these two scenarios.
The new setting could result in this:
20 work processes in priv mode since about 45 minutes.
20 other work processes in priv mode since about 30 minutes.
All the other work processes in priv mode since about 15 minutes.
So nobody else will be allowed to work. This will only change in 15 minutes, when those exceeding 60 minutes will be cancelled.
On the other hand, if know know for sure that only a few of your users will go priv, your suggestion will work.
There are a lot of SAP notes or help texts about priv mode. You might want to look for different solutions. Or it might be necessary to buy more memory for your SAP server.
regards
Maybe you are looking for
-
I used to have a greek flag on my top bar, but when i updated to 10.6.4 i no longer have it. How do I get it back?
-
my skype name nasadam is taken someone.. and also change my email address as well..
-
Grid in ACR7 lens correction?
i'm sorry if this is a stupid question, but is there any way to turn on a grid in the ACR lens correction?
-
Every time I import a scanned image into iphoto and try to use it iphoto crashes. I upgraded from 3 to 4 with the same results. Does not seem to have a problem with images from camera only scanned.
-
How do I get my iPhone 4 contacts switched onto my iPhone 3G
I need to switch my iPhone 4 contacts over to my iPhone 3G. I can't get ICloud to work on the iPhone 3G and the AT&T store couldn't get the contacts switched over either. How can I switch them over without using iCloud, the store, or apple support ov