Hi. My Name Jinsik KIM.
Net.TeamPolicyUpDelay Setting Does Not Working For ESXi 6.7
The application environment is like this.
[Version]
VxRAIL : 4.7.300
vCenter : 6.7.0.40000
ESXi : 6.7.0, 14320388
vSAN : 6.7.0 14328953
[Question]
In a configuration that works with Big Switch Cloud Fabric,
In accordance with the recommendation, the Net.TeamPolicyUpdelay setting was changed from the default value of 100ms to 30000ms.
After redundancy link is restored, it should have 30000ms of delay but is applied as 100ms, causing ping loss. (Configuration that does not cause ping loss when having linkage with BCF)
In ESXi 6.5, Net.TeamPolicyUpdelay works as set.
In ESXi 6.7, even if you changed the setting, you had to have a delay (30000 ms) at the time you set, but it did not work.
[Log] - /var/log/vmkernel.log
1. ESXi 6.5 (Net.TeamPolicyUpdelay In-Value 30000ms)
==Link Up ===
2019-11-14T05:13:40.639Z cpu34:66155)i40en: i40en_UpdateUplinkLinkStatus:3956: Link is up at 10000 Mbps for device vmnic0 (0x43076800f040)
2019-11-14T05:13:40.639Z cpu34:66155)netschedHClk: NetSchedHClkNotify:2900: vmnic0: link up notification
2019-11-14T05:13:40.794Z cpu26:80280)netschedHClk: NetSchedHClkWatchdogSysWorld:4552: vmnic0: link up event received, device running at 10000 Mbps so setting queue depth to 86460 bytes with expected 1310 bytes/us
2019-11-14T05:13:40.794Z cpu26:80280)WARNING: netschedHClk: NetSchedHClkWatchdogSysWorld:4695: vmnic0: failed to push the coalescing settings cranking upinflight window to infinite: Not supported
2019-11-14T05:13:42.197Z cpu32:66191)i40en: i40en_UpdateUplinkLinkStatus:3956: Link is up at 10000 Mbps for device vmnic2 (0x43076adef040)
2019-11-14T05:13:42.655Z cpu32:66191)i40en: i40en_UpdateUplinkLinkStatus:3956: Link is down for device vmnic2 (0x43076adef040)
2019-11-14T05:13:43.131Z cpu32:66191)i40en: i40en_UpdateUplinkLinkStatus:3956: Link is up at 10000 Mbps for device vmnic2 (0x43076adef040)
===Link Delay and Resume ===
2019-11-14T05:14:10.794Z cpu35:65677)netschedHClk: NetSchedHClkPortQuiesce:4980: vmnic1: received a force quiesce for port 0x3000010
2019-11-14T05:14:10.794Z cpu35:65677)NetPort: 1881: disabled port 0x3000010
2019-11-14T05:14:10.794Z cpu35:65677)NetLB: 2237: Driver claims supporting 7 RX queues, and 7 queues are accepted.
2019-11-14T05:14:10.794Z cpu35:65677)NetLB: 2233: Driver claims supporting 7 TX queues, and 7 queues are accepted.
2019-11-14T05:14:10.794Z cpu35:65677)NetPort: 3208: resuming traffic on DV port 10
2019-11-14T05:14:10.794Z cpu35:65677)netschedHClk: NetSchedHClkNotify:2900: vmnic0: link up notification
2019-11-14T05:14:10.794Z cpu35:65677)Uplink: 10095: enabled port 0x3000010 with mac e4:43:4b:94:b1:f0
2. ESXi 6.7 (Net.TeamPolicyUpdelay In-Value 30000ms)
2019-11-14T05:18:58.738Z cpu26:2097902)i40en: i40en_UpdateUplinkLinkStatus:3910: Link is up at 10000 Mbps for device vmnic1 (0x4307cf1ab0c0)
2019-11-14T05:18:58.738Z cpu26:2097902)netschedHClk: NetSchedHClkNotify:2952: vmnic1: link up notification
2019-11-14T05:18:59.515Z cpu40:2369236)netschedHClk: NetSchedHClkWatchdogSysWorld:4632: vmnic1: link up event received, device running at 10000 Mbps so setting queue depth to 86460 bytes with expected 1310 bytes/us
2019-11-14T05:18:59.515Z cpu40:2369236)WARNING: netschedHClk: NetSchedHClkWatchdogSysWorld:4770: vmnic1: failed to get the coalescing settings: Not supported
2019-11-14T05:18:59.515Z cpu40:2369236)WARNING: netschedHClk: NetSchedHClkWatchdogSysWorld:4784: vmnic1: failed to push the coalescing settings cranking upinflight window to infinite: Not supported
2019-11-14T05:18:59.515Z cpu40:2097340)NetqueueBal: 5032: vmnic1: device Up notification, reset logical space needed
2019-11-14T05:18:59.515Z cpu40:2097340)netschedHClk: NetSchedHClkPortQuiesce:5046: vmnic0: received a force quiesce for port 0x2000002
2019-11-14T05:18:59.515Z cpu40:2097340)NetPort: 1580: disabled port 0x2000002
2019-11-14T05:18:59.516Z cpu40:2097340)NetPort: 2680: resuming traffic on DV port 5
2019-11-14T05:18:59.516Z cpu40:2097340)netschedHClk: NetSchedHClkConnect:5180: vmnic1: multiple queue not enabled
2019-11-14T05:18:59.516Z cpu40:2097340)netschedHClk: NetSchedHClkNotify:2952: vmnic1: link up notification
2019-11-14T05:18:59.516Z cpu40:2097340)Uplink: 11689: enabled port 0x2000002 with mac e4:43:4b:94:c2:b1
[Capture]
I think it's a bug in esxi 6.7 u3 version.
Does anyone have the same problem with me?
Thankyou.