Why maintain separate kernels for ide / scsi?

I don't really understand why arch has separate kernels for ide and scsi. I was just reading some forum posts about difficulties in installing arch, and they are all related to SATA. Apparently, you need to boot the CD with the non-default scsi kernel in order to get sata drives recognized for installation.
Also, people are expecting to be using their USB storage devices, which all require SCSI, but apparently this isn't well known, judging by the number of posts asking for help on this matter.
I hope we can discuss this issue here, and hopefully make changes for the Wombat install CD.
The SCSI kernel is only a ~600kb larger download than the default IDE, so I don't think the reason for separating them is based on size. Is the kernel bootup time difference really noticable on an i686 system? (ie, people running arch don't have such old computers where it would be a big deal)
If there are legitimate reasons for keeping an ide-only kernel, then I suggest we at least make SCSI the default, and provide kernel26-noscsi as an option for those who need it. But I would personally prefer to get rid of the ide-only kernel alltogether.

paranoos wrote:Also, people are expecting to be using their USB storage devices, which all require SCSI, but apparently this isn't well known, judging by the number of posts asking for help on this matter.
USB storage devices work very well with the IDE kernel just by loading a few kernel modules (2 as far as I remember) or use hotplug. It worked out of the box for me. Read the Wiki for this.
About the SATA installation problem, maybe a Wiki page (or some lines in the online installation guide) about this would be useful. I plan buying a SATA HDD (with a new PC) soon and I am still thinking wether Arch will install on it without no problem.

Similar Messages

Maintain separate table for these fields

I want to maintain separate table for this fields.
1. Sales order number
2.item number
3.material
4.qty pricing date price change date if u user changes price manually.
5. Order price
6.sales value
Where will we get these fields from which table.

Hi,
1. Sales order number = VBAK-VBELN
2.item number = VBAP-POSNR
3.material = VBAP-MATNR
4.qty pricing date price change date if u user changes price manually.
Qty from VBAP-KWMENG
Price date = VBKD-PRSDT
Change Date = VBAK-AEDAT
5. Order price = VBAK-NETWR
Thanks,
Mandar

Update kernel steps for IDES 4.7 - Windows - MS SQL SERVER 2000

Hello,
Concerning updating the kernel - for IDES 4.7, Windows, SQL SERVER 2000 - is there a way to locate exactly the.SAR files - for the database dependent & database independent parts - which are compatible with the used SAP version.
Also, does updating the kernel involve only the following:
- Downloading the .SAR file.
- Taking a backup copy of the exe/run folder's content.
- Stopping the SAP system and all related services.
- Unpacking the .SAR files using the SAPCAR utitilty,
i.e. are there any specific steps to be done after the unpacking ?
Thanks in advance.
Reda Khalifa
IT Department - Almansour Automotive Group - Egypt

Hi Reda,
I guess You want to upgrade kernel from 6.20 to 6.40.
Go to service.sap.com -> downloads -> SAP Support Packages -> Search for SP and Upgrades -> enter "kernel 6.40" and follow the link "kernel 6.40 32-BIT".
On next window choose from location :SAP KERNEL 6.40 \ Windows Server \ Database independent following SARs (title):
-      enq
-      disp+work package
-      dsrlib
-      enserver
-      librfc
-       MMC (optional)
-      Kernel Part I (for Basis 640/620/610) Q1/2007 (very important)
-       saposcol
From location:SAP KERNEL 6.40 \ Windows Server \ MS SQL Server choose following SARs:
-      Database Library
-      R3LOAD
-      R3load Control
-       Kernel Part II (for Basis 640/620/610) Q1/2007 (very important)
Unpack it to newKernel directory and copy it to \exe\run directory.
Next step Frank Deijkers already wrote.
On the end of upgrade go to \exe\run catalog and run sapstartsrv.exe program for registering new service and COM library.
Regards,
Marcin

Git - maintain two branches for separate kernel26 packages?

Hi, I'm maintaining two kernel26 packages on the AUR (kernel26-ice and kernel26-rt-ice). I'm wondering whether git would be suitable for maintaining a repo for both packages.
kernel26-rt-ice is just derivative from kernel26-ice, but is currently very very left behind since the -rt patch devs haven't released since 2.6.33. This may or may not change in future, but the resulting difference between the PKGBUILDs increases my maintainance time.
Previously I just had a small script which copied the kernel26-ice files to another folder, some sed hackery to change names/configs etc. Now I'm wondering if I could set up git branches to do the same. My idea is as follows:-
1. Two branches named kernel26-ice and kernel26-rt-ice
2. Changes made in kernel26-ice (the more active one) can be cherry-picked into kernel26-rt-ice. This includes general cleaning up of PKGBUILD etc. but not the commits which involve updating variables etc.
Is this task suitable for git? Or should I just maintain separate trees and diff them every once in a while?

ngoonee wrote:
Dieter@be wrote:
yes we can!
git >> manual trees and patches
What is the benefit over manual trees/patches in this specific case? I understand primarily the source control benefit (using it myself for my thesis software development), but am unfamiliar on whether it would actually make my life easier maintaining both packages.
err there are so many things I could go into..
with git you will basically have no issues.
if you currently have no issues, don't switch. if you currently have one or more issues with your workflow, switch to git and those issues will be gone.

SCSI Emulation for IDE Tape Devices

Linux has a nifty driver called ide-scsi.o, an SCSI emulation for IDE devices, that allows SCSI commands to sent to IDE devices. These IDE devices (CDROM, tape drives, etc...) appear as SCSI devices.
Does Solaris have something similiar?
Thanks,
Quang

legal wrote:
I have tried burning without setting scsi emulation but give me some errors, so how can I set scsi emulation for burning CD/DVD on k3b???
It's in rc.conf by adding "ide-scsi"???
I'm using Kernel 2.6.4 (IDE version), KDE 3.2.1 and K3b 0.11.6.
Note: I've tried to upgrade k3b from 0.11.6 to 0.11.7 with "pacman -Su k3b" but say to me that already in last version - 0.11.6.
with kernel 2.6 you don't need anymore scsi emulation for cd burning, i do that, on the bootloader ( lilo is what i use) i add at the append line this --> hdc=ide-cd , and that's all, maybe even that is not needed, not sure. If you want to test if that works open a shell and type --> cdrecord --dev=ATAPI --scanbus
It should show your burner.
For k3b you maybe need to run the setup program first, and make sure your cdrdao is uptodate.

Why Dev Guide says "create a separate project for each ADF task flow ..."

Hi,
We have a BPM process application which is having a few human tasks. The form for each of the human tasks is similar. We’d like to be able to share some page fragments or code between them and create single web application to handle all the human tasks for this BPM process application. Then I found the following in the Oracle document “Developer’s Guide for Oracle SOA Suite 11g Release 1 (11.1.1.4.0)”:
“If the SOA composite contains multiple human tasks, create a separate project for each ADF task flow associated with each human task.”
I wonder if we have to do that and if so, what is the reason.
Do you have any clue? Please help.
Thanks a lot,
Helen

Hi Helen
1. I do not know the reason why they mentioned like separate project for each human task. NO, this is not required and we can have just one UI Project like MyTaskForms and have all the JSPs in this same project for each Human Task.
2. On our side, we have like 10 Human Tasks, all using the same Payload. So we have 10 .task file. For the very first .task file, we used Auto Generate Project and gave project name like OurAppTaskForms. The reason we did this is, so that it adds all the required dependent libraries, tag libs etc properly. For all remaining 9 .tasks file, we did NOT use auto generate. Instead we used standard New -> JSF From Human Task and select one task at a time and follow the steps. These steps are covered in the online documentation. At the end in our BPM App, we have just one UI Project with all the 10 screens.
3. ALSO, all the 10 screens have some common data. This we put in a reusable JSF Fragment page. And associated a custom Managed Bean for this to set and get values from this fragment and also some Actions for some buttons in the JSF. Then we included this JSF in each taskform jsf file, uisng the include tag. We added this include tag, somewhere in the top. So all screens share common jsf fragment page.
4. We can refine the above stuff further. As per above, we still have 10 JSF screens for each one of the 10 .tasks. Ofcourse all these 10 jsf have same common jsf fragment. BUT instead of 10 individual jsf screens, we can have only ONE Jsf screen also. Refer online docs for the details. The challenging part is, since we have only one jsf, there may be some bussiness stuff, validatoin stuff to take care of in custom ValidationCallback Handler classes. Say for Task1, we had to show some buttons, but NOT for Task2. So we had to handle this logic. Also in validation call back, we can always get each Tasks, title also. So we can workaround this issue also. We are in the process of doing this right now. But first 3, we already did and they are working fine.
My personal suggestion is first go with single jsf fragment and include in all tasks jsp files. Any changes to common elements need to make only in one place i.e fragment page.
Thanks
Ravi Jegga

Why do we have separate apps for phone, messages, and contacts?

I'm trying to reduce the number of apps that I need to run, and this one just boggles my mind.
Why do we have separate apps for phone, messages, and contacts?
The messages is to tightly coupled to the phone, that it doesn't make any sense for them to be separate.
The contacts app is nothing more then looking at the contacts from the phone app.
What's the story here?
Does it have something to do with the contacts also being part of email app?
Actually, that's my other biff... why are they entangled as such?
Phone contacts should have phone#, email contacts should have email addresses.
If you want a contact to have both phone# and email address, then the two apps (phone/email) should be combined.

The contacts app is nothing more then looking at the contacts from the phone app.
What's the story here?"
The forums would be full of people complaining that you had to go through an additional step ( opening the phone app) to get to contacts. I use the contacts app from the home screen far more often then I use it in relation to calling someone.
I like the way it is far more than the way you propose.
You can leave feedback for Apple at:
http://www.apple.com/feedback

My Apple TV and IPad drop after x mintues from "Home Sharing". To restart it, I need to close iTunes on my Windows 7 PC, open it , then I can listen again X minutes (15-90 minutes) to my music via "Home Sharing". Any idea why? What worked for you?

My Apple TV and IPad drop after x mintues from "Home Sharing". To restart it, I need to close iTunes on my Windows 7 PC, open it , then I can listen again X minutes (15-90 minutes) to my music via "Home Sharing". Any idea why? What worked for you?

I believe if you updated your Apple TV, you need iTunes 11 to make it work.

Why are albums that have various artists appearing as separate albums for each track?

why are albums that have various artists appearing as separate albums for each track?

Check the tags via the info panel on each system.
You may see that one version has settings for 'Disk x of y' disks.
There are other fields that should be compared too, like 'grouping'. You also need to be certain the playlist (or the Library) are sorted by the same criteria e.g. by 'album' is different to 'album by artist' etc. That can cause unexpected differences in the way it displays.
I suspect match has given you iTunes copies of the files with Apple's own metadata whilst your originals are from CD with your custom tags.

I have Creative Cloud and just downloaded Ae... but it is version CS6... not CC2014.1 . Any idea why the download is for an older version?

I have Creative Cloud and just downloaded Ae... but it is version CS6... not CC2014.1 . Any idea why the download is for an older version?

If there are problems with updating or with the permissions then easiest is to download the full version and trash the currently installed version to do a clean install of the new version.
Download a new copy of the Firefox program and save the disk image (dmg) file to the desktop
* Firefox 7.0.x: http://www.mozilla.com/en-US/firefox/all.html
* Trash the current Firefox application to do a clean (re-)install
* Install the new version that you have downloaded
Your profile data is stored elsewhere in the Firefox Profile Folder, so you won't lose your bookmarks and other personal data if you uninstall and (re)install Firefox.
* http://kb.mozillazine.org/Profile_folder_-_Firefox

Why is iCloud asking for apple id password? I have started using Last Pass to generate passwords that are more secure. Ideas please?

Why is iCloud asking for apple id password? I have started using Last Pass to generate passwords that are more secure. Ideas please?
GCG

The iCloud password isn't saved anywhere on your device, this is the way it's supposed to work.

Why - Separate Partnerlinks for same child process

Early in our development, we noticed that we had to have separate partnerlinks for the same service if two (or more) parallel paths were concurrently invoking the service receive. Our guess was the delivery of the child Async send required separate partnerlinks to deliver the sent message to the correct parent process flow. I would be very interested in hearing from any one's experience if that is the correct conclusion.
However, I now have a serial flow of BPEL processes, as illustrated below, where the Async send response was lost in step A3.
Process A
    |
   Step A1
     +----> Async Process B Request
     |                         |
     +<----- Async Process B Send Done response
     |
     |
   Step A2
     +----> Async Process B Request
     |                         |
     +<----- Async Process B Send Done response
     |
     |
    Step A3
     +----> Async Process B Request
     |                         |
     +<----- Async Process B Send Done response
     | All of the above process B requests and wait for received response are through the same partnerlink in the A process. There is no activity in the A process between the Send to B and the following Receive from B. The time between the B Request and the B send response can be long since it can involve a human invoked response.
I have a case where in Step A3 the "Async Process B Send Done response" in the B flow is completed, but the receive in the A3 is still waiting.
*(Question 1)* Is there a problem using the same partnerlink in A for all of these process B send and receive pairs?
*(Question 2)* Is there another explanation for the missing response to the A3 step?

Can you explain how to perform the manual recovery? Is it from the BPELConsole's "BPEL Processes" tab and "Perform Manual Recovery" button? There are zero messages in that screen.
Edited by: BCurtis on Apr 30, 2009 10:51 AM

Why are there two separate packages for extra/zip and extra/unzip?

The package info for both indicate that they come from the same website. It seems strange and inconsistent with what is normally expected. For example, there are not separate packages for gzip and gunzip or bzip2 and bunzip2.

1. They are represented as distinct programs on the website.
2. They are developed independently, they even have different version numbers.
3. Although the binaries are very small, some people only need one of them. I never create zip files, but I have to unzip them often.
4. They seem to have different dependencies.

Maintaining multiple universes for separate environments

I know this is a basic question, but are there guidelines on how to manage separate universes for different versions of the same physical database structure? For example, what's a good way to manage universes for separate test environments where the table owner/qualifier is different for each environment (i.e. test1.tablename, test2.tablename, test3.tablename, prod.tablename).

There are a couple of options that I will share with you.
Option one is the least work on the BOBJ side and the most work on the DBA side. The DBA must create an account with access only to the required tables (or views). This means that account does not require the qualifier (test1.table_name) and instead can reference only by table_name. Now there is no issue when migrating from test1 to test2 to prod.
Unfortunately, our DBA team didn't like that option. So I created a short VBA script, hosted in Excel, that we use to update our universe as we move it from one environment to another. We've been using it for years now, and it has been working fine. However, starting with BOBJ 4.0 they removed the option of VBA / SDK scripting from the universe, so I'm not sure what we're going to do when we get to that version.
If you're interested in having a look at the script, I've posted it on my blog here:
http://www.dagira.com/2008/03/06/using-the-designer-sdk-to-ease-migrations/

Ide-scsi & supermount doesn't work

hi again,
i have still some things not running,
my actual questions are:
why isn't my ide-scsi emulation for cdrecord working?
i added 'hdc=autotune,ide-scsi' in the append line, but nothing happens... :cry: (yes i did a 'lilo' after that) could the reason be, that i take the lilo from my still installed mdk-system?
how do i enable supermount? or is it not included in the standard-kernel? i added the lines for supermount in /etc/fstab but mounting it says 'no filesystem supermount' or so... :?:
thx again 4 ur help!!!!!
SiToR (writing from [email protected])

Hi, I'm using SUPERMOUNT with the kernel patches from: http://kernel.kolivas.org and Kernel-2.4.22. It works great.
Be sure to add the following lines to /etc/devfsd.conf, if you want normal users to have access to sound and to be able to format floppies using kfloppy:
REGISTER sound/.* PERMISSIONS root.users 660
REGISTER snd/.* PERMISSIONS root.users 660
REGISTER floppy/.* PERMISSIONS root.users 660
Here's my /etc/fstab:
/dev/discs/disc0/part3 / reiserfs defaults 0 0
/dev/discs/disc0/part2 swap swap defaults 0 0
none /proc proc defaults 0 0
none /dev/shm tmpfs defaults 0 0
tmpfs /tmp tmpfs size=2G,nr_inodes=10k 0 0
/dev/discs/disc0/part1 /mnt/windows auto umask=0 0 0
/dev/cdroms/cdrom0 /mnt/cdrom supermount dev=/
dev/cdroms/cdrom0,--,user,exec,dev,suid,noauto,unhide 0 0
/dev/floppy/0 /mnt/floppy supermount dev=/dev/floppy/0,--,user,exec,dev,suid,noauto,umask=0 0 0

Why maintain separate kernels for ide / scsi?

Similar Messages

Maybe you are looking for