OVM 3.0.1 local repository problem

Good morning all, i am really new in OVM and i am facing a big issue that stops me evaluating this product.
I have a couple of servers, connected to a S.A.N. array. I can see from both the servers i added to a clustered pool, and i am able to create a shared repository without problems.
I am not able to see local disks in the OVM manager administration and therefore i can't create local repositories. I tried all i found in this forum, but without success.
Let's focus on server1: it has a couple of 146GB disks. I used one of them for OVS installation leaving the second disk alone, without partitioning it.
Tried to create local repository in the clustered pool, but no way...
So i created a single full-disk partition and retried to create repo: still no way
Then i created an ocfs2 filesystem in the new partition but, again, i couldnt see physical local server1 disk.
Every time i changed partitions configuration, i obviously did rescanning of physical disks.
I all my tests, local physical disks selection list in Generic Local Storage Array @ node1 is always empty.
Any hint about solving this issue? Any good pointer to an hands-on guide (official docs are not so good)? Any suggestion about what to look at in log files for debugging?
Any answer is welcome...
Thank you all!

I was able to do this as follows
1. have an untouched unformatted disk (no partitions, no file system)
2. in hardware under the vmserver name , scan for the disk and it should show in the list
3. in the repository section of home, add the repository as physical disk
4. "present" (green up and down arrows) the physical disk on the vmserver itself (dont ask me why you have to do this but if you dont it wont find its own disk)

Similar Messages

OVM 2.2 and NFS Repository Problems

Hi All.
I have recently started trying to upgrade our installation to 2.2
but have run into a few problems, mostly relating to the different
way that storage repositories are handled in comparison to 2.1.5.
We use NFS here to provide shared storage to the pools.
I wanted to setup a new two node server pool (with HA), so I upgraded
one of the servers to from 2.1.5 to 2.2 to act as pool master. That
worked ok and this server seems to be working fine in isolation:
master# /opt/ovs-agent-2.3/utils/repos.py -l
[ * ] 865a2e52-db29-48f1-98a0-98f985b3065c => augustus:/vol/OVS_pv_vpn
master# df /OVS
Filesystem 1K-blocks Used Available Use% Mounted on
augustus:/vol/OVS_pv_vpn
47185920 16083008 31102912 35% /var/ovs/mount/865A2E52DB2948F198A098F985B3065C
(I then successfully launched a VM on it.)
The problem is when I try to add a second server to the pool. I did
a fresh install of 2.2 and configured the storage repository to be the
same as that used on the first node:
vm1# /opt/ovs-agent-2.3/utils/repos.py --new augustus:/vol/OVS_pv_vpn
vm1# /opt/ovs-agent-2.3/utils/repos.py -r 865a2e52-db29-48f1-98a0-98f985b3065c
vm1# /opt/ovs-agent-2.3/utils/repos.py -l
[ R ] 865a2e52-db29-48f1-98a0-98f985b3065c => augustus:/vol/OVS_pv_vpn
When I try to add this server into the pool using the management GUI, I get
this error:
OVM-1011 Oracle VM Server 172.22.36.24 operation HA Check Prerequisite failed: failed:<Exception: ha_precheck_storage_mount failed:<Exception: /OVS must be mounted.> .
Running "repos.py -i" yields:
Cluster not available.
Seems like a chicken and egg problem: I can't add the server to the pool without a
mounted /OVS, but mounting /OVS is done by adding it to the pool? Or do I have that
wrong?
More generally, I'm a bit confused at how the repositories are
supposed to be managed under 2.2.
For exaple, the /etc/init.d/ovsrepositories script is still present,
but is it still used? When I run it, it prints a couple of errors and
doesn't seem to mount anything:
vm1# service ovsrepositories start
/etc/ovs/repositories does not exist
Starting OVS Storage Repository Mounter...
/etc/init.d/ovsrepositories: line 111: /etc/ovs/repositories: No such file or directory
/etc/init.d/ovsrepositories: line 111: /etc/ovs/repositories: No such file or directory
OVS Storage Repository Mounter Startup: [ OK ]
Should this service be turned off? It seem that ovs-agent now takes
responsibility for mounting the repositories.
As an aside, my Manager is still running 2.1.5 - is that part of the
problem here? Is it safe to upgrade the manager to 2.2 while I still
have a couple of pools running 2.1.5 servers?
Thanks in adavance,
Robert.

rns wrote:
Seems like a chicken and egg problem: I can't add the server to the pool without a
mounted /OVS, but mounting /OVS is done by adding it to the pool? Or do I have that
wrong?You have that wrong -- the /OVS mount point is created by ovs-agent while the server is added to the pool. You just need access to the shared storage.
For exaple, the /etc/init.d/ovsrepositories script is still present,
but is it still used?No, it is not. ovs-agent now handles the storage repositories.
As an aside, my Manager is still running 2.1.5 - is that part of the
problem here? Yes. You absolutely need to upgrade your Manager first to 2.2 before attempting to create/manage a 2.2-based pool. The 2.1.5 Manager doesn't know how to tell the ovs-agent how to create/join a pool properly. The upgrade process is detailed in [the ULN FAQ|https://linux.oracle.com/uln_faq.html#10].

//solved // pacman local repository problem / testing vs testing

i have a serious problem with pacman, i hope someone can help me out.
i've tried to create a local repo and made a really stupid mistake by calling it 'testing' (i used repo-add)
now if i try to update the system with -Syu I get a flood of warnings for all my packages being newer than platform core etc
there is obviously a conflict with the usual 'testing' and I couldn't find a way to revert this. repo-remove didn't help either.
is there a way to fix it?
tia
Last edited by JYQ (2013-04-04 10:02:58)

parazyd wrote:Try editing your pacman.conf and remove it.
I tried it already. Didn't help.
jasonwryan, thanks but i have a little experience with pacman. the problem is a confict in the database.
manpages do not cover this.
"Creating & Modifying Packages( Discussion and support regarding building packages with makepkg and ABS)
was a natural choice for the thread because there are people who do the same thing.
Last edited by JYQ (2013-04-04 10:23:30)

[SOLVED] Create local repository

Hello everyone!
First of all, I'd like to apologize for the terminology. I'm still not sure about what are the names I should use. That said, here's my problem: I want to set up a local repository to place all the packages I've ever built with makepkg. The main reason that has motivated me to do so is that, when building a package, I often need to install make dependencies, such as, for example, xorg-server-devel. That package in particular comes from xorg-server, which I have built by myself. Thus, if a package required it, I should install the correspondent package (the one I've built), and not the generic one coming from the repositories (this case may not be a good example, but, well, just think about kernel headers).
In order to set up the repository, I read Allan McRae's articles about makepkg, repo-add and package signing. I generated a key with GPG as explained here, taking the defaults, and added it to the pacman's keyring:
gpg --export Kalrish > Kalrish.key
pacman-key -a Kalrish.key
I also edited /etc/makepkg.conf as follows:
BUILDENV=(fakeroot !distcc color !ccache check sign)
Note that I've enabled sign.
I decided to maintain my "local repository" (please, correct me if I'm not making use of the correct terminology) in /var/abs/local. For that purpose, I edited /etc/makepkg.conf:
$ grep -m1 PKGDEST /etc/makepkg.conf
PKGDEST=/var/abs/local
I then built dhcpcd (which is relatively small), just to try out. makepkg put the package file in /var/abs/local as I intended; then, I did this:
repo-add 'local.db.tar.gz' 'dhcpcd-5.6.8-3-x86_64.pkg.tar.xz'
It asked me for the password to unlock my secret key. After entering it, all went well:
==> Adding package 'dhcpcd-5.6.8-3-x86_64.pkg.tar.xz'
-> Adding package signature...
-> Computing checksums...
-> Creating 'desc' db entry...
-> Creating 'depends' db entry...
==> Creating updated database file 'local.db.tar.gz'
And finally added the repo in /etc/pacman.conf (I'm only showing the relevant part of the file):
[local]
Server = file:///var/abs/local
I updated the package databases just fine with pacman -Syy. But, after removing the previously-downloaded packages from /var/cache/pacman/pkg, when trying to install the package:
$ pacman -Sy dhcpcd
:: Synchronizing package databases...
local is up to date
core is up to date
extra is up to date
community is up to date
multilib is up to date
:: dhcpcd is in IgnorePkg/IgnoreGroup. Install anyway? [Y/n] y
warning: dhcpcd-5.6.8-3 is up to date -- reinstalling
resolving dependencies...
looking for inter-conflicts...
Packages (1): dhcpcd-5.6.8-3
Total Installed Size: 0.23 MiB
Net Upgrade Size: 0.00 MiB
:: Proceed with installation? [Y/n] y
(1/1) checking keys in keyring [######################] 100%
(1/1) checking package integrity [######################] 100%
error: dhcpcd: signature from "Kalrish (Kalrish Bäakjen) <[email protected]>" is invalid
:: File /var/cache/pacman/pkg/dhcpcd-5.6.8-3-x86_64.pkg.tar.xz is corrupted (invalid or corrupted package (PGP signature)).
Do you want to delete it? [Y/n] n
error: failed to commit transaction (invalid or corrupted package (PGP signature))
Errors occurred, no packages were upgraded.
So it seems that I have not added my key properly. Can anybody please help me? Any answer, even if it explained why I shouldn't be trying to do this or why I should try other alternatives, will be welcomed.
Best regards,
Kalrish
Last edited by Kalrish (2013-06-28 18:26:30)

drcouzelis wrote:... if you built the package on that machine yourself then all of the dependencies are already installed.
I was talking about make dependencies, not runtime dependencies. I always build with makepkg -sr :)
drcouzelis wrote:I hope I'm not being too critical. I just don't want you to work harder than you need to. :)
Of course not, don't worry.
Kalrish wrote:Any answer, even if it explained why I shouldn't be trying to do this or why I should try other alternatives, will be welcomed.
I love thought-provoking answers as well as personal opinions. Who knows when an unexpected explanation may change my mind and make me think different? You're right: it may not be worth it. But, well, you yourself pointed the "elegant" reason out:
drcouzelis wrote:I think the only benefit of creating a local repository is so you can use "pacman -S" instead of "pacman -U".
If pacman found the packages by itself, I wouldn't have to execute pacman -U --asdeps for every make dependency that is required by a package - but rather execute makepkg -sr and simply let pacman pick those packages I've built by myself from my local repository, and the rest from the official repos.

Subversion: Cannot import files into a local repository

If following the directions from here:
http://www.oracle.com/technology/products/jdev/htdocs/partners/addins/exchange/subversion/subversion.html
I've got the local repository created and can successfully connect to it, but when it try to import files is get a popup saying "Unable to import source files into the Subversion repository at svn://localhost."
This is the stack trace:
org.tigris.subversion.svnclientadapter.SVNClientException: org.tigris.subversion.javahl.ClientException: svn: 170001: Authorization failed
     at org.tigris.subversion.svnclientadapter.javahl.AbstractJhlClientAdapter.doImport(AbstractJhlClientAdapter.java:700)
     at oracle.jdevimpl.vcs.svn.imp.SVNImportWizard.executeImport(SVNImportWizard.java:476)
     at oracle.jdevimpl.vcs.svn.imp.SVNImportWizard.commitWizard(SVNImportWizard.java:204)
     at oracle.jdevimpl.vcs.svn.imp.SVNImportWizard.mav$commitWizard(SVNImportWizard.java:78)
     at oracle.jdevimpl.vcs.svn.imp.SVNImportWizard$3$1.run(SVNImportWizard.java:137)
     at java.lang.Thread.run(Thread.java:595)
Caused by: org.tigris.subversion.javahl.ClientException: svn: 170001: Authorization failed
     at org.tigris.subversion.javahl.JavaHLObjectFactory.throwException(JavaHLObjectFactory.java:435)
     at org.tmatesoft.svn.core.javahl.SVNClientImpl.throwException(SVNClientImpl.java:1224)
     at org.tmatesoft.svn.core.javahl.SVNClientImpl.doImport(SVNClientImpl.java:601)
     at org.tigris.subversion.svnclientadapter.javahl.AbstractJhlClientAdapter.doImport(AbstractJhlClientAdapter.java:696)
     ... 5 more
I'm using JDeveloper 10.1.3.0.4.3673 and Subversion 1.4.6. Please advise.

Thank you so much for your help. With some deeper googling, i found some info on editing the svnserve.conf file. That fixed my authentication issue. But alas, there is a new problem: Same popup, new stack trace:
org.tigris.subversion.svnclientadapter.SVNClientException: org.tigris.subversion.javahl.ClientException: svn: Malformed network data
     at org.tigris.subversion.svnclientadapter.javahl.AbstractJhlClientAdapter.doImport(AbstractJhlClientAdapter.java:700)
     at oracle.jdevimpl.vcs.svn.imp.SVNImportWizard.executeImport(SVNImportWizard.java:476)
     at oracle.jdevimpl.vcs.svn.imp.SVNImportWizard.commitWizard(SVNImportWizard.java:204)
     at oracle.jdevimpl.vcs.svn.imp.SVNImportWizard.mav$commitWizard(SVNImportWizard.java:78)
     at oracle.jdevimpl.vcs.svn.imp.SVNImportWizard$3$1.run(SVNImportWizard.java:137)
     at java.lang.Thread.run(Thread.java:595)
Caused by: org.tigris.subversion.javahl.ClientException: svn: Malformed network data
     at org.tigris.subversion.javahl.JavaHLObjectFactory.throwException(JavaHLObjectFactory.java:435)
     at org.tmatesoft.svn.core.javahl.SVNClientImpl.throwException(SVNClientImpl.java:1224)
     at org.tmatesoft.svn.core.javahl.SVNClientImpl.doImport(SVNClientImpl.java:601)
     at org.tigris.subversion.svnclientadapter.javahl.AbstractJhlClientAdapter.doImport(AbstractJhlClientAdapter.java:696)
     ... 5 more
I've been looking all morning for solutions, but to no avail.

System update configuration issue with local repository related to machine type

Hi,
Sorry if the issue has been posted earlier, i tried searching but cudnt find relevant post. Here is my issue:
My machine is Thinkpad R52. Type as printed on label is: 1858-CLG
While adding machine type in update retriever, I added machine type as 1858 (as it doesnt accept 1858CLG). After connecting to the support site, it downloaded all required updates onto a local(shared) repository.
I installed system update, configured it to pickup update from local (network) repository in the registry.
Now when I launch system update, it determines my machine type as 1858CLG & it connects to local repository only to display message as "No updates were found for this machine type"
Though I cud see updates if I directly connect to help center.
I think the problem is with machine type identification, not sure though.
Please advice on what I'm doing wrong. Is it with update retriever configuration or system update configuration.
Thanks much,

Solution for me was to change status of packages from "Test" to "Active" using Update Retriever.
Since this is not documented in Best Practices (but is stated in chapter 5 - modify settings) I spent several hours figuring it out..
-brian

Oracle VM 3.1.1 and local repository

Hi there,
I have created one LUN hardware RAID 10 for 4x300G SAS disks, and then installed Oracle VM Server 3.1.1, during the installation I determined the remaining space on sda for Oracle VM repository on partition sda2 mounted on directory /OVS OCFS2 file system, installation completed successfully, and then I installed Oracle VM Manager 3.1.1 on a separate machine it is installed and working fine.
Now, Oracle VM Manager discover the Oracle VM server smoothly but unable to detect that remaining space on /OVS, my question how to use this remaining space for Oracle VM repository as I read on the new release document "Release Notes" : said that this issue is fixed with this release and mentioned it as bug in their previous version. I wish if any ideas or workaround to fix this issue as I always hope, there is a manual intervention to solve this issue and want at the same time to keep performance benefit of RIAD 10 configuration.
Thanks in advance,

Sorry, you get to do it over.
You must install the Local Repository on an unused, complete, unpartitioned LUN of it's own. Build about a 50G LUN for Oracle VM and and additional LUN for the remaining terabyte. OVM will then see the space and utilize the repo. Create a non-clustered Server Pool.
You can put OVM in much smaller space, but I like to have "extra" (add a /dev/sda4 and mkfs on it) stuff on my OS disks. If you don't want to do that, OVM fits nicely in 5G. I"d give it 10G for growth with later upgrades, however.

Local Repository Concept in BODS 4.0

Hi Experts - On the BODS 4.0 logon screen I am using enterprise user name and password to Logon and After that logon got successful it shows the list of repositories. For this your repository must be registered in CMC.
Now if any developer want to use his local repository , created on this system locally , how he can use this ?
Or we need to create local repository for each developer on the Database on which we have created default installation repository and then need to register them in CMC.
Please suggest.
Thanks
R

Hi Rohit,
Just to make sure we are talking about the same thing.
When I say "local repository" I mean this as opposed to "central repository" which in reality is a different kind of repository and used to administer different versions of objects. A revision control mechanism.
I have the feeling that when you are referring to local repositories you are looking at repositories on the laptop/machine of the developer:
You will have to access the databases on the developer laptops through the CMS server running to authenticate.
Adding them is no problem, the database has to be reachable by the CMS machine and you simply put the laptop-ip/port and other login information like username/password... into the CMS. This will create an entry that will show up when the developer is logging in to the CMS server and then a connection to the database will be built by the client on the developer's laptop.
As Seb mentioned it normally does not help to have local repositories on your laptop other than if you need access to the database.
You can only get the benefits of the authentication mechanism like a centralized and more granular access control through the CMS, which would not work if you could ALSO login to the database directly...
Look at it like a link: You have to login to the CMS to get the link to the database. From the authentication onward the communication is between client and database again.
Norbert

Error add local repository[SOLVED]

I create a local repository as wiki says. http://wiki.archlinux.org/index.php/Cus … nd_gensync.
My db.tar.gz is at
/home/lq/abs/myrepo.db.tar.gz
And here is my pacman.conf
# /etc/pacman.conf
# See the pacman.conf(5) manpage for option and repository directives
# GENERAL OPTIONS
[options]
# The following paths are commented out with their default values listed.
# If you wish to use different paths, uncomment and update the paths.
#RootDir = /
#DBPath = /var/lib/pacman/
#CacheDir = /var/cache/pacman/pkg/
#LogFile = /var/log/pacman.log
#HoldPkg = pacman glibc
# If upgrades are available for these packages they will be asked for first
SyncFirst = pacman
XferCommand = /usr/bin/wget --passive-ftp -c -O %o %u
#XferCommand = /usr/bin/curl %u > %o
#CleanMethod = KeepInstalled
# Pacman won't upgrade packages listed in IgnorePkg and members of IgnoreGroup
#IgnorePkg =xorg-server xorg-server-utils xf86-input-evdev xf86-input- synaptics kernel26 kernel26-firmware libgl ati-dri kernel-headers glibc binutils gcc-libs gcc
#IgnoreGroup =
#NoUpgrade =
#NoExtract =
# Misc options (all disabled by default)
#NoPassiveFtp
#UseSyslog
#ShowSize
#UseDelta
#TotalDownload
# REPOSITORIES
# - can be defined here or included from another file
# - pacman will search repositories in the order defined here
# - local/custom mirrors can be added here or in separate files
# - repositories listed first will take precedence when packages
# have identical names, regardless of version number
# - URLs will have $repo replaced by the name of the current repo
# Repository entries are of the format:
# [repo-name]
# Server = ServerName
# Include = IncludePath
# The header [repo-name] is crucial - it must be present and
# uncommented to enable the repo.
# Testing is disabled by default. To enable, uncomment the following
# two lines. You can add preferred servers immediately after the header,
# and they will be used before the default mirrors.
#[testing]
#Include = /etc/pacman.d/mirrorlist
[core]
# Add your preferred servers here, they will be used first
Include = /etc/pacman.d/mirrorlist
[extra]
# Add your preferred servers here, they will be used first
Include = /etc/pacman.d/mirrorlist [community]
# Add your preferred servers here, they will be used first
Include = /etc/pacman.d/mirrorlist
[myrepo]
Server = file:///home/lq/abs
# An example of a custom package repository. See the pacman manpage for
# tips on creating your own repositories.
#[archlinuxfr]
#Server = http://repo.archlinux.fr/i686
#[Gooth]
#Server = http://public.gooth.cn/archlinux/gooth/os/i686
And when update,it says
2009-10-12 21:26:59 (9.92 MB/s) - `/var/lib/pacman/community.db.tar.gz.part' saved [377225/377225]
file:///home/lq/abs/myrepo.db.tar.gz: Unsupported scheme `file'.
:: Starting full system upgrade...
local database is up to date
Could anyone tell me what is the problem ?
Last edited by fbchaos (2009-10-12 14:56:01)

Looks like you are using wget as your download agent (check XferCommand in pacman.conf). wget doesn't understand file, you can either switch to curl or the internal downloader.
Last edited by Garns (2009-10-12 14:31:18)

Best way to upgrade the local repository from 11.7.3 to 12.2.3.3 DS 3

Hi, we are going to upgrade our DI 11.7.3 to DS3 - 12.2.3.3 in Sep. I found out there are different ways to upgrade the local repository from the SAP doc and tech support. I try to find a best and safe way to do it. If anyone had experince and inputs on this. It would be great appreciated. It seems there are these options:
we all agree that need to back up the repostiry db, and export the repository to atl file.
Option 1:
Upgrade the repository during the Data Services installation. But the tech support said there are times where this does not work as expected and Option 2 below needs to be implemented.
Option 2:
a) Export the local repository, via 11.7.3 Designer, to an ATL file.
b) Uninstall 11.7.3.
c) Create a new database in SQL Server to use with 12.2.3.3. (Optionally, use the existing 11.7.3 database. If you do, make a backup of it before making any changes).
d) Install Data Services 12.2.3.3 and reference the database created in step "c" above during installation (select the Create option).
e) Log in to the new local repo, via 12.2.3.3 Designer, and import the ATL that was exported in step "a" above.
Option 3:
a. Create a new database u2013 20 MB space
b.Create a new repo in 11.7.3 point to ds_local via Repository Manager (click create)
c.Log in 11.7 designer and Repository import from out.atl
d.Install 12.2.3.3 without upgrde the repostory
e.use epository manager in 12.2.3 to upgrade the new repository

Hi.
If you import the 11.7 atl into a 12.2 Designer you will face a warning message like this:
"The version 11.7 of the ATL file <c:\export.atl> is lower than the repository version <12.2>.
To update an ATL file, import it into a repository of the same version then ugrade that repository"
So, the correct method to upgrade the Data Integrator 11.7 repository to Data Services 12.2.x is the following:
- Export the complete Data Integrator 11.7 repository to ATL.
- Create a NEW Data Integrator 11.7 Local Repository.
- Import the ATL created in step 1 to the NEW Data Integrator 11.7 Repository. At this point you have two identical local repositories.
- From the Data Services 12.2.x system, open Repository Manager.
- Enter the Database connection information to the NEW repository created in step 2.
- Select UPGRADE, to upgrade this repository to support Data Services 12.2.x.
- Log into Data Services Designer to start using this Local Repository.
Regards,
Leo.

Previous Version of Local Repository

Hi ALL.
I need some help....
Want to find out if there is a way of retrieving my local repository's work?
The issue I face is that I mistakenly used the "Get Lastest Version" (from the Central Repository) feature & as a result it has overwritten my work in my local repository.
Is there a way to revert back?? I know the one way would have been to have created an .ATL file.
You help will be very much appreciated.
Thanks.

Hi Jeff.
They probably have.
Nevertheless, I found that most of the data flows are published in the Central Repo.
So what I did, is the the logs from a previously run job to see which dataflows were used, and I basically then re-used the same data flows, as a result the re-work was not as much.
Anyways moral of the story, ALWAYS make ATL back-ups.
Thanks guys for your help!

Get message " The local repository database could not be found or is corrupted"

Hi
I want to update the System drivers , but constanly getting the massage
"The local repository database could not be found or is corrupted"
not sure what to do
cannot find any reference to the error
Please assist
This is getting to me
Johan

Hi,
I have the same issue (on a windows 2008 R2 SP1)
I want to do a backup from my automation (a simple batch .cmd)
I'm doing this command:
C:\Users\Administrator>wbadmin start backup -backuptarget:s:\backup -exclude:s: -allCritical -include:C:,R: -quiet
wbadmin 1.0 - Backup command-line tool
(C) Copyright 2004 Microsoft Corp.
Retrieving volume information...
This will back up volume System Reserved (100.00 MB),New Volume(R:),DISK_S(S:) (Selected Files),Local Disk(C:) to s:\backup.
ERROR - The specified backup location could not be found or is not a supported backup storage location.
The only way to get this working, is working via UNC to the local drive S-drive
C:\Users\Administrator>wbadmin start backup -backuptarget:\\myserver\s$\backup -exclude:s: -allCritical -include:C:,R: -quiet
However, the backup succeeds with these warnings:
Retrieving volume information...
This will back up volume System Reserved (100.00 MB),New Volume(R:),DISK_S(S:) (Selected Files),Local Disk(C:) to
\\myserver\s$\backup
and giving a warning
The backup of volume DISK_S(S:) could not be completed. Error: The requested operation could not be completed due to a file system limitation
The backup operation completed with errors.
Looks like the wbadmin does not read the -exclude option?
Regards,
David

Local Repository Question...

First - I really appreciate all the work that everyone has put into this, I'm really starting to like this distro a lot...
Second...
Still pretty new to the linux world. Messed with Xubuntu for the last few months, but I wanted to learn more... So I've ended up with Arch...
I've done a couple of installs to a GUI with some success, but still have a few things to get figured out...
Third - My question...
I spend a lot of time in places where I do not have internet access and I really like to mess, so I've download all of the files from an extra repository to set up a local repo on my hd.
I ran the perl script quickhack from the wiki against the directory, copied the file the script created to the directory with all of the other files and renamed it to "current.db.tar.gz"
From there I set my pacman.conf to point to the directory and done a sync/refresh and everything works fine...
I can search/install any of the files that are listed, but when I try to do "pacman -S xfce4" or "pacman -S gnome" I get the "not found in sync db"... I can install any of the separate files but not an entire group...
What is it that I'm overlooking ?
Thanks...

sparrow wrote:
I spend a lot of time in places where I do not have internet access and I really like to mess, so I've download all of the files from an extra repository to set up a local repo on my hd.
I ran the perl script quickhack from the wiki against the directory, copied the file the script created to the directory with all of the other files and renamed it to "current.db.tar.gz"
From there I set my pacman.conf to point to the directory and done a sync/refresh and everything works fine...
Why did you do that? There are few ways to create local mirror (full or only installed packages) of Arch repos, all described on Wiki.
You don't need to merge Arch repos into one local repository.
sparrow wrote:
I can search/install any of the files that are listed, but when I try to do "pacman -S xfce4" or "pacman -S gnome" I get the "not found in sync db"... I can install any of the separate files but not an entire group...
What is it that I'm overlooking ?
That's because you regenerated repo db with this script, so you lost your groups.
If I understand you correctly, you need to look at these articles:
http://wiki.archlinux.org/index.php/Local_Mirror
http://wiki.archlinux.org/index.php/How … me_Network

Local repository installation

I have read all the wiki articles and also found this post:
http://bbs.archlinux.org/viewtopic.php?id=30431
For some reason I cannot install form the local repository in /va/cache/pacman/pkg for the life of me!
Every time I try and install something, like KDE-mod it constantly tries to connect to the internet and download the pacage although it is in /var/cache/pacman/pkg:
[root@carley carley]# pacman -S kdemod-complete
:: group kdemod-complete:
kdemod-arts kdemod-kde-common kdemod-kdeaccessibility kdemod-kdeaddons-base
kdemod-kdeaddons-knewsticker-scripts kdemod-kdeaddons-ksig
kdemod-theme kdemod-ui-kdemod qt-enhanced
:: Install whole content? [Y/n] y
resolving dependencies... done.
looking for inter-conflicts... done.
Targets: kdemod-kde-common-3.5.7-7 qt-enhanced-3.3.8-11 kdemod-arts-1.5.7-5
kdemod-kdelibs-3.5.7-13 hwinfo-13.28-1 kdemod-ui-kdemod-1.0-1
kdemod-kdebase-3.5.7-17 kdemod-kdeaccessibility-3.5.7-2
kdemod-kdepim-base-3.5.7-4 kdemod-kdemultimedia-base-3.5.7-4
kdemod-kdeaddons-base-3.5.7-4 kdemod-kdeaddons-knewsticker-scripts-3.5.7-2
Total Package Size: 276.07 MB
Total Installed Size: 527.28 MB
Proceed with installation? [Y/n] y
:: Retrieving packages from kdemod...
error: failed retrieving file 'kdemod-kdelibs-3.5.7-13-i686.pkg.tar.gz' from kdemod.ath.cx : Transient resolver failure
error: failed retrieving file 'kdemod-ui-kdemod-1.0-1-i686.pkg.tar.gz' from kdemod.ath.cx : Transient resolver failure
error: failed retrieving file 'kdemod-kdebase-3.5.7-17-i686.pkg.tar.gz' from kdemod.ath.cx : Transient resolver failure
error: failed retrieving file 'kdemod-kdepim-base-3.5.7-4-i686.pkg.tar.gz' from kdemod.ath.cx : Transient resolver failure
error: failed retrieving file 'kdemod-kdepim-akregator-3.5.7-3-i686.pkg.tar.gz' from kdemod.ath.cx : Transient resolver failure
error: failed retrieving file 'kdemod-kdevelop-3.4.1-3-i686.pkg.tar.gz' from kdemod.ath.cx : Transient resolver failure
warning: failed to retrieve some files from kdemod
error: failed to commit transaction (unexpected error)
pacman keeps trying to connect to the internet to install!
Is there an option I should pass to force it to install from the local cache?
I can install with pacman -U /package but this will take ages for all the kde-mod packages.
Last edited by MW (2007-08-15 02:28:58)

Could you paste the above pacman output with --debug flag ?
And also maybe output of ls -l /var/cache/pacman/pkg

Local Repository User Rights for the Data Designer

Hello,
we are using the Data Services with 4 users and one local repository stored on the server.
Is it possible to hide in the Data Designer foreign objects and projects from other users in the local repository? What's the best practice for that, for every user a seperate repository or to restrict the repository rights with the cmc?
Our system:
Windows Server 2008
Data Services 4.0 SP1
IBM DB2 9.7 (Repository)
Thanks for some tips and hints!

Hi Alexander,
At any point in time, you should only have one user connected to one repository in Designer.
If two users are using the same repository at the same time, then there is a risk of corruption of repository objects.
Best practice is to have a repository for each (concurrent) developer and a central repository for version management.
Depending how testing is organised, you might also need to have repositories for unit test / QA (UAT) and finally a production repository.
For good measure, an additional central reposotiry for production can be valuable:
- if development for new project start before the 'previous' project has gone into live.
- You can prepare go-live by checking in all the objects required for live into central-production first.
- It's the final safety measure, if objects in production are corrupted and dev/test systems have modified versions.
So, a typical landscape would look like this:
DEVELOPMENT:
n number of development repositories, one for each developer
1 test repository (local)
1 dev central repository
TEST
1 UAT repository (local)
PROD
1 PROD repository (local)
1 central repository
Jan.

OVM 3.0.1 local repository problem

Similar Messages

Maybe you are looking for