Description of problem: Currently the sync_conduit.get_units scopes all the units specific to the repo its initialized. But for errata case where an erratum can span across multiple repos, we need to check if it exists in a broader scope during the sync. Otherwise with the way we have it today, the previously synced errata from another repo gets overridden.
I need some more details on the expectation here. - Will you be editing units retrieved from this call? - Will you be creating an association from units retrieved from this into the repository being syncced? I'm also not sure I understand "Otherwise with the way we have it today, the previously synced errata from another repo gets overridden.", but that depends on your answers above. If you're going to edit the erratum retrieved from this call, that's what's going to happen anyway. If the call is purely going to be read only, this becomes a non-issue.
build: 0.331
v1 bug with verification steps https://bugzilla.redhat.com/show_bug.cgi?id=860800
verified [root@preethi-el6-pulp ~]# rpm -q pulp-rpm-server pulp-rpm-server-2.0.6-0.2.beta.noarch [root@preethi-el6-pulp ~]# [root@preethi-el6-pulp ~]# pulp-admin rpm repo content errata --repo-id rhel-6-w --repo-id rhel-6-w-optional --erratum-id RHBA-2012:1305 +----------------------------------------------------------------------+ Erratum: RHBA-2012:1305 +----------------------------------------------------------------------+ Id: RHBA-2012:1305 Title: openswan bug fix update Summary: Updated openswan packages that fix a bug are now available for Red Hat Enterprise Linux 6. Description: Openswan is a free implementation of IPsec (Internet Protocol Security) and IKE(Internet Key Exchange) for Linux. The openswan packages contain daemons anduser-space tools for setting up Openswan. It supports the NETKEY/XFRM IPseckernel stack that exists in the default Linux kernel. Openswan 2.6 and lateralso supports IKEv2 (Internet Key Exchange Protocol version 2), which is definedin RFC5996. This update fixes the following bug: * When a tunnel was established between two IPsec hosts (say host1 and host2)utilizing DPD (Dead Peer Detection), and if host2 went offline while host1continued to transmit data, host1 continually queued multiple phase 2 requestsafter the DPD action. When host2 came back online, the stack of pending phase 2requests was established, leaving a new IPsec SA (Security Association), and alarge group of extra SA's that consumed system resources and eventually expired.This update ensures that openswan has just a single pending phase 2 requestduring the time that host2 is down, and when host2 comes back up, only a singlenew IPsec SA is established, thus preventing this bug. (BZ#852454) All users of openswan are advised to upgrade to these updated packages, whichfix this bug. Severity: Type: bugfix Issued: 2012-09-24 00:00:00 Updated: 2012-09-24 00:00:00 Version: 1 Release: Status: final Reboot Suggested: No Updated Packages: openswan-0:2.6.32-19.el6_3.i686 References: ID: None Type: self Link: https://rhn.redhat.com/errata/RHBA-2012-1305.html [root@preethi-el6-pulp ~]#
Pulp v2.0 released