Hide Forgot
Created attachment 1131821 [details] Log files Description of problem: Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. https://beaker.engineering.redhat.com/jobs/1243679 2. http://lab-02.rhts.eng.bos.redhat.com/beaker/logs/recipes/2525+/2525607/console.log Found below FAILED info in both server and client console.log. Bringing up interface ocrdma_roce.43: Bringing up interface ocrdma_roce.43: 802.1Q VLAN Support v1.8 Ben Greear <greearb> All bugs added by David S. Miller <davem> 8021q: adding VLAN 0 to HW filter on device ocrdma_roce Determining IP information for ocrdma_roce.43... Determining IP information for ocrdma_roce.43... failed. failed. [FAILED] [FAILED] Bringing up interface ocrdma_roce.45: Bringing up interface ocrdma_roce.45: Determining IP information for ocrdma_roce.45... Determining IP information for ocrdma_roce.45... failed. [FAILED] failed. [FAILED] [root@rdma-qe-04 ~]$ ip addr 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: ocrdma_roce: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9000 qdisc mq state UP qlen 1000 link/ether 00:90:fa:29:48:6a brd ff:ff:ff:ff:ff:ff inet 172.31.40.4/24 brd 172.31.40.255 scope global ocrdma_roce inet6 fe80::290:faff:fe29:486a/64 scope link valid_lft forever preferred_lft forever 3: ocrdma_10g_2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 00:90:fa:29:48:72 brd ff:ff:ff:ff:ff:ff 4: tg3_1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000 link/ether 9c:b6:54:bb:48:84 brd ff:ff:ff:ff:ff:ff inet 10.16.41.226/21 brd 10.16.47.255 scope global tg3_1 inet6 2620:52:0:102f:9eb6:54ff:febb:4884/64 scope global dynamic valid_lft 2591890sec preferred_lft 604690sec inet6 fe80::9eb6:54ff:febb:4884/64 scope link valid_lft forever preferred_lft forever 5: tg3_2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 9c:b6:54:bb:48:85 brd ff:ff:ff:ff:ff:ff 6: ocrdma_roce.43@ocrdma_roce: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9000 qdisc noqueue state UP link/ether 00:90:fa:29:48:6a brd ff:ff:ff:ff:ff:ff inet6 fe80::290:faff:fe29:486a/64 scope link valid_lft forever preferred_lft forever 7: ocrdma_roce.45@ocrdma_roce: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9000 qdisc noqueue state UP link/ether 00:90:fa:29:48:6a brd ff:ff:ff:ff:ff:ff inet6 fe80::290:faff:fe29:486a/64 scope link valid_lft forever preferred_lft forever 3. Actual results: Expected results: Additional info:
Hi developers, Would you like to help take a look? Thanks.
you attached an rds-tools-qib.log file. I referenced the beaker log file to see relevant errors -- I see ocrdma_roce fail, but I also see rds stack traces in the log as well. Could you (re-)test with the -616 kernel. 4 patches were added to -617 for ocrdma, and I just want to eliminate those 4 separate patches as the source of the errors. Also, did this test pass in 6.7 ?
I had already told all of you about this issue. The DHCP/VLAN/REORDER_HDR combination is the culprit. 1) comment VLAN_ID=43, it will take more than 80 seconds for roce.43 to get dhcp IP. 4 out of 5 succeed to get IP. 2) comment REORDER_HDR=0 The first round of "ifup ocrdma_roce.43" got IP in 5 second. However, the second round hang on and never exit. Ctrl+C killed it after 13 minutes. If you ifdown/ifup the vlan interface in a for loop, dhcp will not work at all with different error messages. So, the temporary workaround is to comment REORDER_HDR=0, and never abuse the vlan interface with ifup/ifdown.
(In reply to Honggang LI from comment #3) > I had already told all of you about this issue. > > The DHCP/VLAN/REORDER_HDR combination is the culprit. > > 1) comment VLAN_ID=43, it will take more than 80 seconds for roce.43 > to get dhcp IP. 4 out of 5 succeed to get IP. > > 2) comment REORDER_HDR=0 > The first round of "ifup ocrdma_roce.43" got IP in 5 second. However, > the second round hang on and never exit. Ctrl+C killed it after 13 minutes. > If you ifdown/ifup the vlan interface in a for loop, dhcp will not work > at all with different error messages. > > So, the temporary workaround is to comment REORDER_HDR=0, and never > abuse the vlan interface with ifup/ifdown. I digged out the email Honggang sent before. We need "uncomment REORDER_HDR=0", and "uncomment VLAN_ID=43" in the /etc/sysconfig/network-scripts/ifcfg-ocrdma_roce.XX. I just tried this workaround and it works.
Red Hat Enterprise Linux 6 is in the Production 3 Phase. During the Production 3 Phase, Critical impact Security Advisories (RHSAs) and selected Urgent Priority Bug Fix Advisories (RHBAs) may be released as they become available. The official life cycle policy can be reviewed here: http://redhat.com/rhel/lifecycle This issue does not meet the inclusion criteria for the Production 3 Phase and will be marked as CLOSED/WONTFIX. If this remains a critical requirement, please contact Red Hat Customer Support to request a re-evaluation of the issue, citing a clear business justification. Note that a strong business justification will be required for re-evaluation. Red Hat Customer Support can be contacted via the Red Hat Customer Portal at the following URL: https://access.redhat.com/