PDA

View Full Version : I/O error: Can't provide /repodata/repomd.xml



plaubscher
29-Jan-2019, 14:12
To whom it may concern

I am trying to register five SLES 10 SP3 clients using SUSE Manager 3.2.I am getting the under mentioned error messages, when I execute the bootstrap script to the clients.I did recreate the SLES 10 SP3 activation key, but still getting the under-mentioned error messages.

May somebody please assist me.

I/O error: Can't provide /repodata/repomd.xml from https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
Entering non-interactive mode.
Entering 'no-gpg-checks' mode.
Determining https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap source type...
https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap is type YUM
Not found: Media Exception
Not found: Media Exception
Added Installation Sources:
[x]* susemanager:bootstrap (https://cen-prd-susemngr/pub/repositories/sle/10/3/bootstrap)
Entering non-interactive mode.
Entering 'no-gpg-checks' mode.
Refreshing susemanager:bootstrap
URI: https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
Please insert media [Download (curl) error for 'https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap/repodata/repomd.xml':
Error code: Unrecognized error
Error message: error:0D0C50A1:asn1 encoding routines:ASN1_item_verify:unknown message digest algorithm
] # 1 and type 'y' to continue or 'n' to cancel the operation. [y/n]: n
(A)bort, (R)etry, (I)gnore? a
All system sources have been refreshed.
I/O error: Can't provide /repodata/repomd.xml from https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
I/O error: Can't provide /repodata/repomd.xml from https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
Error while refreshing the source: Downloading metadata failed (is YUM source?) or user did not accept remote source. Aborting refresh.Entering non-interactive mode.
Entering 'no-gpg-checks' mode.
Restoring system sources...
Please insert media [Download (curl) error for 'https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap/repodata/repomd.xml':
Error code: Unrecognized error
Error message: error:0D0C50A1:asn1 encoding routines:ASN1_item_verify:unknown message digest algorithm
] # 1 and type 'y' to continue or 'n' to cancel the operation. [y/n]: n
(A)bort, (R)etry, (I)gnore? a
I/O error: Can't provide /repodata/repomd.xml from https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
I/O error: Can't provide /repodata/repomd.xml from https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
Failed to restore sources
package spacewalk-check is not installed
ERROR: Failed to install all missing packages.

malcolmlewis
29-Jan-2019, 15:08
Hi
Wow 10 SP3.... did you use the createrepo command to generate the repo?

kwk
30-Jan-2019, 08:10
Did you create the bootstrap repository as outlined in the documentation ?

plaubscher
31-Jan-2019, 14:46
Malcolm

Thanks for the reply, sorry for only responding now. Yes I did used the createrepo command.I did try the client troubleshooting TID https://www.suse.com/support/kb/doc/?id=7023241, and restarted the server.I want to delete and recreate my activation key as well, over the weekend.

I will come back to you.

Regards

plaubscher
05-Feb-2019, 18:50
Yes, I did follow the correct procedure for creating a bootstrap.

For the preparing of JeOS I did the following, which I think could be the problem.

1. With the customers' code I did register successfully to SCC.

2.I added the under-mentioned SUSE-Manager repositories.

3. To verify the repositories, I used the command SUSEConnect - -LIST extensions
The following showed up :

- sle-module-adv-systems-management/12/x86_64 -r `xxxx
- sle-module-web-scripting/12/x86_64
.- sle-module-toolchain/12/x86_64 -r xxxx
- PackageHub/12.3/x86_64
- PackageHub/12.3/x86_64 -r xxxx
- suse-openstack-cloud-crowbar/8/x86_64 -r xxxx
- sle-module-containers/12/x86_64 -r xxxx
- sle-module-hpc/12/x86_64 -r xxxx
- hpe-helion-openstack/8/x86_64 -r xxxx
- sle-module-legacy/12/x86_64 -r xxxx
- sle-module-public-cloud/12/x86_64 -r xxxx
-sle-module-cap-tools/12/x86_64 -r xxxx
- ses/5/x86_64 -r xxxx
- sle-ha/12.3/x86_64 -r -r xxxx
- sle-ha-geo/12.3/x86_64 -r xxxx
- sle-ha/12.3/x86_64 -r xxxx
- SUSE-Manager-Proxy/3.2/x86_64 xxxx
- PackageHub/12.3/x86_64 xxxx
- SUSE Manager Server 3.2 x86_64 ( Installed )

Do I really need all these repositories except for SUSE Manager Server 3.2 x86_64, sle-ha/12.3/x86_64 -r -r , sle-ha-geo/12.3/x86_64 -r and sle-ha/12.3/x86_64 -r xxxxx ?

plaubscher
05-Feb-2019, 18:59
What also could be the problem is that I created my filesystem using XFS and not BTRFS.Do I need to leave the filesystem as default when doing the installation ? My current filesystem layout is undermentioned.

df -h
Filesystem Size Used Avail Use% Mounted on
devtmpfs 16G 0 16G 0% /dev
tmpfs 16G 120K 16G 1% /dev/shm
tmpfs 16G 634M 16G 4% /run
tmpfs 16G 0 16G 0% /sys/fs/cgroup
/dev/sda3 66G 38G 29G 58% /
/dev/sdb1 200G 33M 200G 1% /var/lib/pqsql
/dev/sdc1 900G 189G 712G 21% /var/spacewalk
/dev/sda2 2.0G 67M 1.8G 4% /boot
tmpfs 3.2G 16K 3.2G 1% /run/user/479
tmpfs 3.2G 0 3.2G 0% /run/user/0

There is no data in /var/lib/pqsql, which is strange. Is it best to reinstall the server with the default BTRFS filesystem as proposed ?

malcolmlewis
05-Feb-2019, 21:09
Yes, I did follow the correct procedure for creating a bootstrap.

For the preparing of JeOS I did the following, which I think could be the problem.

1. With the customers' code I did register successfully to SCC.

2.I added the under-mentioned SUSE-Manager repositories.

3. To verify the repositories, I used the command SUSEConnect - -LIST extensions
The following showed up :

<snip>

Do I really need all these repositories except for SUSE Manager Server 3.2 x86_64, sle-ha/12.3/x86_64 -r -r , sle-ha-geo/12.3/x86_64 -r and sle-ha/12.3/x86_64 -r xxxxx ?
Hi
No, I only have;



zypper lr -E

# | Alias | Name | Enabled | GPG Check | Refresh
---+----------------------------------------------------------------+---------------------------------+---------+-----------+--------
4 | SUSE_Linux_Enterprise_Server_12_SP3_x86_64:SLES12-SP3-Pool | SLES12-SP3-Pool | Yes | (r ) Yes | No
6 | SUSE_Linux_Enterprise_Server_12_SP3_x86_64:SLES12-SP3-Updates | SLES12-SP3-Updates | Yes | (r ) Yes | Yes
9 | SUSE_Manager_Server_3.2_x86_64:SUSE-Manager-Server-3.2-Pool | SUSE-Manager-Server-3.2-Pool | Yes | (r ) Yes | No
11 | SUSE_Manager_Server_3.2_x86_64:SUSE-Manager-Server-3.2-Updates | SUSE-Manager-Server-3.2-Updates | Yes | (r ) Yes | Yes

malcolmlewis
05-Feb-2019, 21:18
What also could be the problem is that I created my filesystem using XFS and not BTRFS.Do I need to leave the filesystem as default when doing the installation ? My current filesystem layout is undermentioned.

df -h
Filesystem Size Used Avail Use% Mounted on
devtmpfs 16G 0 16G 0% /dev
tmpfs 16G 120K 16G 1% /dev/shm
tmpfs 16G 634M 16G 4% /run
tmpfs 16G 0 16G 0% /sys/fs/cgroup
/dev/sda3 66G 38G 29G 58% /
/dev/sdb1 200G 33M 200G 1% /var/lib/pqsql
/dev/sdc1 900G 189G 712G 21% /var/spacewalk
/dev/sda2 2.0G 67M 1.8G 4% /boot
tmpfs 3.2G 16K 3.2G 1% /run/user/479
tmpfs 3.2G 0 3.2G 0% /run/user/0

There is no data in /var/lib/pqsql, which is strange. Is it best to reinstall the server with the default BTRFS filesystem as proposed ?
There should be more usage of that directory....



lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 698.7G 0 disk
├─sda1 8:1 0 500M 0 part /boot
├─sda2 8:2 0 1K 0 part
├─sda5 8:5 0 100G 0 part /
├─sda6 8:6 0 6G 0 part [SWAP]
├─sda7 8:7 0 100G 0 part /var/lib/pgsql
└─sda8 8:8 0 492.1G 0 part /var/spacewalk

df -h
Filesystem Size Used Avail Use% Mounted on
devtmpfs 2.8G 0 2.8G 0% /dev
tmpfs 2.8G 24K 2.8G 1% /dev/shm
tmpfs 2.8G 35M 2.8G 2% /run
tmpfs 2.8G 0 2.8G 0% /sys/fs/cgroup
/dev/sda5 100G 5.5G 94G 6% /
/dev/sda5 100G 5.5G 94G 6% /var/opt
/dev/sda5 100G 5.5G 94G 6% /opt
/dev/sda5 100G 5.5G 94G 6% /var/lib/machines
/dev/sda5 100G 5.5G 94G 6% /var/lib/mysql
/dev/sda5 100G 5.5G 94G 6% /var/lib/libvirt/images
/dev/sda5 100G 5.5G 94G 6% /var/log
/dev/sda5 100G 5.5G 94G 6% /var/tmp
/dev/sda5 100G 5.5G 94G 6% /var/lib/mariadb
/dev/sda5 100G 5.5G 94G 6% /usr/local
/dev/sda5 100G 5.5G 94G 6% /home
/dev/sda5 100G 5.5G 94G 6% /var/spool
/dev/sda5 100G 5.5G 94G 6% /var/lib/named
/dev/sda5 100G 5.5G 94G 6% /var/lib/mailman
/dev/sda5 100G 5.5G 94G 6% /var/cache
/dev/sda5 100G 5.5G 94G 6% /tmp
/dev/sda5 100G 5.5G 94G 6% /var/crash
/dev/sda5 100G 5.5G 94G 6% /srv
/dev/sda1 477M 86M 366M 20% /boot
/dev/sda7 100G 6.6G 94G 7% /var/lib/pgsql
/dev/sda8 492G 28G 465G 6% /var/spacewalk
tmpfs 573M 0 573M 0% /run/user/0


I have SLED 15 x86_64 and SLES 15 x86_64/aarch64 downloaded/synced etc. I plan to add SLE 12 SP4 in the near future.

kwk
06-Feb-2019, 13:25
Yes, I did follow the correct procedure for creating a bootstrap.

Then make sure that the following directory path exists:
/srv/www/htdocs/pub/repositories/sle/
and below you have <sle-version>/<service-pack> directories like
15/0
for SLES 15 SP0.

Now check that a "bootstrap" directory exists and is populated

> ls /srv/www/htdocs/pub/repositories/sle/15/0/bootstrap
noarch repodata x86_64

If anything is missing, run "mgr-create-bootstrap-repo" again.



For the preparing of JeOS I did the following, which I think could be the problem.

1. With the customers' code I did register successfully to SCC.


Now I'm confused. Registration to SCC is only needed for the SUSE Manager Server, which I believe you've already set up.

A managed client only needs the SLE "Pool" repository, i.e. from a mounted installation DVD.

Then this client can be bootstrapped to be managed by SUSE Manager.

plaubscher
11-Feb-2019, 13:55
Thanks for the reply

The documentation was a bit confussing for me.I reinstalled the server on vSphere VMWARE not using JeOS.I previously XFS and not BTRFS as proposed.My databse is up and running and I can see and my /var/lib/pgsql is busy and growing.

This is my bootstrap directory for SLES-SP3 :

/srv/www/htdocs/pub/repositories/sle/10/3/bootstrap # ls
total 0
drwxr-xr-x 1 root root 40 Feb 11 13:49 ./
drwxr-xr-x 1 root root 18 Feb 11 05:01 ../
drwxr-xr-x 1 root root 350 Feb 11 05:01 noarch/
drwxr-xr-x 1 root root 370 Feb 11 13:49 repodata/
drwxr-xr-x 1 root root 1692 Feb 11 05:01 x86_64/

The problem is when running the bootstrap for SLES 10-SP3.

cat bootstrap-sles10-sp3.sh \ | ssh root@server_ipaddress /bin/bash - I still get the follwing error message

URI: https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
Please insert media [Download (curl) error for 'https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap/repodata/repomd.xml':
Error code: Unrecognized error
Error message: error:0D0C50A1:asn1 encoding routines:ASN1_item_verify:unknown message digest algorithm
] # 1 and type 'y' to continue or 'n' to cancel the operation. [y/n]: n
(A)bort, (R)etry, (I)gnore? a
All system sources have been refreshed.
I/O error: Can't provide /repodata/repomd.xml from https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
I/O error: Can't provide /repodata/repomd.xml from https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
Error while refreshing the source: Downloading metadata failed (is YUM source?) or user did not accept remote source. Aborting refresh.Entering non-interactive mode.
Entering 'no-gpg-checks' mode.
Restoring system sources...
Please insert media [Download (curl) error for 'https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap/repodata/repomd.xml':
Error code: Unrecognized error
Error message: error:0D0C50A1:asn1 encoding routines:ASN1_item_verify:unknown message digest algorithm
] # 1 and type 'y' to continue or 'n' to cancel the operation. [y/n]: n
(A)bort, (R)etry, (I)gnore? a
I/O error: Can't provide /repodata/repomd.xml from https://cen-prd-susemngr.drdlr.co.za/pub/repositories/sle/10/3/bootstrap
I/O error: Can't provide /repodata/repomd.xml from https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
Failed to restore sources
package spacewalk-check is not installed
ERROR: Failed to install all missing packages.

- When I test the content from web servers using wget I get the follwing error message from my SUSE Manager server.

wget https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
2019-02-11 14:09:01-- https://cen-prd-susemngr.co.za/pub/repositories/sle/10/3/bootstrap
Connecting to 192.168.45:3128... connected.
Proxy tunneling failed: Service UnavailableUnable to establish SSL connection.

- So it seems to be a SSL problem.I used TID 7017147 to verify the symbolic link was added.

cd /etc/pki/trust/anchors
cen-prd-susemngr:/etc/pki/trust/anchors # l
total 4
drwxr-xr-x 1 root root 48 Feb 11 12:07 ./
drwxr-xr-x 1 root root 32 Sep 20 2014 ../
lrwxrwxrwx 1 root root 44 Feb 11 12:07 RHN-ORG-TRUSTED-SSL-CERT -> /srv/www/htdocs/pub/RHN-ORG-TRUSTED-SSL-CERT

- I generated a new RPM with current postscript using these commands:

rhn-ssl-tool --gen-ca --rpm-only --dir="/root/ssl-build"

...working...
Generating CA public certificate RPM:
/root/ssl-build/rhn-org-trusted-ssl-cert-1.0-3.src.rpm
/root/ssl-build/rhn-org-trusted-ssl-cert-1.0-3.noarch.rpm

Make the public CA certificate publically available:
(NOTE: the RHN Satellite or Proxy installers may do this step for you.)
The "noarch" RPM and raw CA certificate can be made publically accessible
by copying it to the /srv/www/htdocs/pub directory of your RHN Satellite or
Proxy server.

- and then used the command

/usr/bin/rhn-deploy-ca-cert.pl --source-dir /root/ssl-build --target-dir /srv/www/htdocs/pub/ --trust-dir=/etc/pki/trust/anchors/

- All seems to be fine running these commands, but still getting "Proxy tunneling failed: Service UnavailableUnable to establish SSL connection" error message.

- I also used TID 7022888 https://www.suse.com/support/kb/doc/?id=7022888, although it is for RHEL 6 servers.

- /usr/bin # update-ca-certificates
- update-ca-certificates force-enabled

- My proxy on the SUSE Manager server is enabled and the firewall is disabled.

kwk
12-Feb-2019, 09:00
Looks like this goes beyond what is supportable via the forum. I'd suggest to open a service request _and_ provide a supportconfig.

jmozdzen
20-Feb-2019, 15:54
Hi *,

could this be a result of the zypper newline issue? I remember having to have the zypper clients updated to access the repositories... and we're talking about *old* clients here ;)

Regards,
J