| Parent Defect ID: | EFA-2900 | Issue ID: | EFA-2900 |
| Severity: | S4 - Low | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | "member vlan add <vlan-range>" and "member bd add <bd-range>" configuration doesnt get removed from the "cluster" instance on the switch, | ||
| Condition: |
1. "member vlan all" and "member bd all" configuration from the "cluster" instance is removed manually on the switch, 2. EPG creation is performed resulting in "member vlan add <vlan-range>" and "member bd add <bd-range>" configuration under the "cluster" instance on the switch. 3. EPG deletion of the EPG created in step 2 is performed. |
||
| Recovery: | "member vlan add <vlan-range>" and "member bd add <bd-range>" configuration can be manually removed from the "cluster" instance on the switch, | ||
| Parent Defect ID: | EFA-3500 | Issue ID: | EFA-3500 |
| Severity: | S4 - Low | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | Request processing takes longer time and CLI remains hung till the request is processed. EPG delete fails after a wait of more than 16 minutes. | ||
| Condition: | Delete EPG having many ( > 256 ) bridge-domain based networks. | ||
| Workaround: |
Incrementally delete networks from the EPG and finally delete EPG Example as below: efa tenant epg update --tenant <> --name <> --operation ctag-range-delete --ctag-range 1001-1256 efa tenant epg delete --tenant <> --name <> |
||
| Parent Defect ID: | EFA-3694 | Issue ID: | EFA-3694 |
| Severity: | S3 - Medium | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.3.0 |
| Symptom: | Request processing takes longer time and CLI remains hung till the request is processed. EPG update may fail if the wait time is more than 16 minutes. | ||
| Condition: |
1. Create EPG with 100 networks. 2. Update EPG with 20 ports/port-channels from one switch using "port-group-add" operation. |
||
| Workaround: | There is no workaround. | ||
| Parent Defect ID: | EFA-3743 | Issue ID: | EFA-3743 |
| Reason Code: | Working as Designed | Severity: | S2 - High |
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.1.0 |
| Symptom: | Config Gen reason CLI does not show the detailed reasons if device is in refreshed state | ||
| Workaround: | EFA fabric show CLI shows these reasons, user can refer to that instead of the debug config-gen reason CLI | ||
| Parent Defect ID: | EFA-3970 | Issue ID: | EFA-3970 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.1.0 |
| Symptom: | EPG create with multiple ctags using a single bridge-domain results in multiple bridge-domain instances to be created on the switch, instead of a single bridge-domain instance on the switch. | ||
| Condition: |
EPG create operation is performed with multiple ctags using a single bridge-domain. Example as below: efa tenant epg create --name vmware31x --tenant t4 --po ncb11 --switchport-mode trunk-no-default-native --ctag-range 300-301 --bridge-domain 300:b2 --bridge-domain 301:b2 |
||
| Workaround: |
EPG update "ctag-range-add" operation can be used to incrementally add multiple ctags using a single bridge-domain. Example as below: efa tenant epg create --name vmware31x --tenant t4 --po ncb11 --switchport-mode trunk-no-default-native --ctag-range 300 --bridge-domain 300:b2 efa tenant epg update --name vmware31x --tenant t4 --operation ctag-range-add --ctag-range 301 --bridge-domain 301:b2 |
||
| Parent Defect ID: | EFA-4109 | Issue ID: | EFA-4109 |
| Severity: | S3 - Medium | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: |
Creation of an L3 EPG with only anycast-ipv4 results in the configuration of both IPv4 and IPv6 address-family under router-bgp. Ideally, the IPv6 address-family should get configured only when the L3 EPG has anycast-ipv6 |
||
| Condition: | Create an L3 EPG with only anycast-ipv4. | ||
| Recovery: | No recovery needed. | ||
| Parent Defect ID: | EFA-4876 | Issue ID: | EFA-4876 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | Bridge domain logical-Interface(s) config under the EFA provisioned port-channel is not reconciled on the switch. | ||
| Condition: |
1. Remove the EFA provisioned port-channel from the switch. 2. Reload the switch in maintenance mode. |
||
| Recovery: |
1. Remove the port-channel interface from the EPGs. 2. Delete and recreate the port-channel interface via EFA. 3. Add the port-channel interface to the EPGs. |
||
| Parent Defect ID: | EFA-5196 | Issue ID: | EFA-5196 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | "cluster-client auto" configuration will not be reconciled for the EFA provisioned port-channel on the switch. | ||
| Condition: |
Configuration drift created on the switch due to scenario as follows: 1. Remove "cluster-client" config from the EFA provisioned port-channel on the switch. 2. Reload the switch in maintenance mode. |
||
| Recovery: |
1. Remove the port-channel interface from the EPGs. 2. Delete and recreate the port-channel interface via EFA. 3. Add the port-channel interface to the EPGs. |
||
| Parent Defect ID: | EFA-5203 | Issue ID: | EFA-5203 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | During device registration, if an error is encountered when installing OAUTH2 cert or HTTPS cert, a warning is generated instead of failing device registration. | ||
| Condition: | This can happen because the system clock in the EFA host and SLX device is not synchronized and the switch will reject the certificate being installed. | ||
| Workaround: | There is no workaround - the device will be registered. | ||
| Recovery: | Ensure that the system time between the EFA host and the devices being used are in sync | ||
| Parent Defect ID: | EFA-5648 | Issue ID: | EFA-5648 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | EPG create with multiple ctags using a single bridge-domain results in multiple bridge-domain instances to be created on the switch, instead of a single bridge-domain instance on the switch. | ||
| Condition: |
EPG create operation is performed with multiple ctags using a single bridge-domain. Example as below: efa tenant epg create --name vmware31x --tenant t4 --po ncb11 --switchport-mode trunk-no-default-native --ctag-range 300-301 --bridge-domain 300:b2 --bridge-domain 301:b2 |
||
| Workaround: |
EPG update "ctag-range-add" operation can be used to incrementally add multiple ctags using a single bridge-domain. Example as below: efa tenant epg create --name vmware31x --tenant t4 --po ncb11 --switchport-mode trunk-no-default-native --ctag-range 300 --bridge-domain 300:b2 efa tenant epg update --name vmware31x --tenant t4 --operation ctag-range-add --ctag-range 301 --bridge-domain 301:b2 |
||
| Parent Defect ID: | EFA-5697 | Issue ID: | EFA-5697 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | EFA provisioned port-channel member interface admin state will not be reconciled on the switch. | ||
| Condition: |
When the switch drifts in configuration wrt intended configuration on the EFA, due to following conditions: 1. EFA provisioned port-channel member interface admin state is changed to shutdown on the switch. 2. Reload the switch in maintenance mode. |
||
| Recovery: |
1. Remove the port-channel interface from the EPGs. 2. Delete and recreate the port-channel interface via EFA. 3. Add the port-channel interface to the EPGs. |
||
| Parent Defect ID: | EFA-5778 | Issue ID: | EFA-5778 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | While using efa firmware-download commands to upgrade switch firmware, intermittently firmware download process gives error "Firmware Activate Failed" for one or more switches | ||
| Condition: | Firmware download process consists of multiple steps, one of which is restart the switch. Upon restart this process tries to connect to the switch and verify if the firmware was successfully finished. Switch reload usually take 1-2 minutes, however very rarely if it take more time, this process waits for switch to restart for 10 minutes. If event after 10 minutes switch restart is not complete, then the it returns this error "Firmware Activate Failed" for that switch | ||
| Workaround: | This is an intermittent issue and there is no workaround to this. | ||
| Recovery: | Start Firmware download process for only the switch with this error from start. | ||
| Parent Defect ID: | EFA-5826 | Issue ID: | EFA-5826 |
| Severity: | S3 - Medium | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | EPG create CLIs can fail with an error "L3VNI is already consumed in the fabric" | ||
| Condition: |
1. Configure EPGs for a tenant. 2. Take the backup of the current configuration of the EFA using "efa show-running-config". 3. Delete tenant with force to clean up all the tenant configuration in EFA and switches. 4. ASCII replay the config backup of "efa show-running-config" taken in step 2. |
||
| Workaround: | Reorder "epg create" CLIs obtained from "efa show-config" in ascending order of l3-vni parameter. | ||
| Parent Defect ID: | EFA-5832 | Issue ID: | EFA-5832 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | "channel-group" config of EFA provisioned port-channel member interface will not be reconciled on the switch. | ||
| Condition: |
When the switch drifts in configuration wrt intended configuration on the EFA, due to following conditions: 1. Remove the "channel-group" config from the EFA provisioned port-channel member interface. 2. Reload the switch in maintenance mode. |
||
| Recovery: |
1. Remove the port-channel interface from the EPGs. 2. Delete and recreate the port-channel interface via EFA. 3. Add the port-channel interface to the EPGs. |
||
| Parent Defect ID: | EFA-5858 | Issue ID: | EFA-5858 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | VRFcreate fails with an error "VRF name <vrf-name> is already taken. Please use different VRF name." | ||
| Condition: | The switch already has a stale or manually created VRF and user is trying to create a VRF (via EFA) with the same name as that of the VRF present on the switch. | ||
| Recovery: | Delete the stale or manually created VRF from the switch before attempting VRF create via EFA. | ||
| Parent Defect ID: | EFA-5869 | Issue ID: | EFA-5869 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | EFA raslog and notification service stops receiving logs from devices/switch which was restored from backed up data. | ||
| Condition: | Occurs after efa restore | ||
| Workaround: | Restart the raslog service using efactl command | ||
| Parent Defect ID: | EFA-5876 | Issue ID: | EFA-5876 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | "cluster member vlan|bd add" and "cluster-client" config reconciliation fails on the switch. | ||
| Condition: |
When the switch drifts in configuration wrt intended configuration on the EFA, due to following conditions: 1. EFA provisioned "cluster" is deleted and recreated with another name on the switch. 2. Reload the switch in maintenance mode. |
||
| Recovery: | Remove and re-add all the port-channels to the EPGs. | ||
| Parent Defect ID: | EFA-5945 | Issue ID: | EFA-5945 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | Few of the EPG's creation takes more time (~20 seconds) compared to the other EPGs. | ||
| Condition: | Create EPGs (with few networks in each EPG) without any delay between EPG create CLIs. | ||
| Workaround: | There is no workaround. | ||
| Parent Defect ID: | EFA-5958 | Issue ID: | EFA-5958 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: |
EPG/Networks create fails with error string as below: failed with netconf error '"0000" is an invalid value.' Where pattern of "0000" can be from one "0" to many "00000"s. |
||
| Condition: |
1. "cluster" instance on the switch doesn't have "member vlan all" and "member bd all" config. 2. Execute EPG create with some ctags. |
||
| Recovery: |
1. Configure "member vlan all" and "member bd all' under "cluster" instance on the switch. 2. Execute inventory update CLI for the switch. 3. Ceate EPG with some ctags. |
||
| Parent Defect ID: | EFA-5960 | Issue ID: | EFA-5960 |
| Severity: | S1 - Critical | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: |
When Switch pumps Raslogs continually (due to faulty SFP) or whatever reason inventory service keeps itself busy in updating device maps. Causing Periodic device discovery not being done. 1.Device discovery will not happen 2.if device is maintenance mode device will not be removed from maintenance mode |
||
| Workaround: | Remove the faulty port or SFP or remove the factor which is causing the switch to pump raslogs continuously . | ||
| Parent Defect ID: | EFA-5965 | Issue ID: | EFA-5965 |
| Severity: | S1 - Critical | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | Firmware download prepare add command fails without giving any error. | ||
| Condition: | When trying to add a device to firmware download, which is part of a non clos fabric, sometimes the mct pairs are configured incorrectly which causes failure of sanity checks in firmware download prepare add command. | ||
| Workaround: | Intermittent issue, no workaround. | ||
| Recovery: | Delete the devices from fabric followed by deleting the fabric. Wait for 2 minutes. Recreate the fabric and then start using the firmware download commands. | ||
| Parent Defect ID: | EFA-5966 | Issue ID: | EFA-5966 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | Further port-group delete operations are not allowed on EPG if there is device connectivity issue while performing port-group-delete operation on EPG | ||
| Condition: |
1. No device connectivity from EFA 2. Run EPG Update to delete port-group 3. Operation failed with "Device connection error" 4. Device connectivity to EFA is restored |
||
| Recovery: |
1. Delete EPG 2. Re-create EPG without port-group which was intended to be removed previously |
||
| Parent Defect ID: | EFA-5978 | Issue ID: | EFA-5978 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | Mac move detect not getting reconciled on device | ||
| Condition: | when mac move limit is set mac move detect does not get reconciled on the device as both mac move limit and mac move detect are together pushed to the device | ||
| Workaround: | Introduce a drift on device by removing mac move limit and then reconcile the device, this would configure both mac move limit and mac move detect. | ||
| Parent Defect ID: | EFA-5987 | Issue ID: | EFA-5987 |
| Severity: | S2 - High | ||
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | If the first IP that shows up from the output of hostname -I isn't the IP to be used for registering the devices with Syslog server then an incorrect IP will be picked up. | ||
| Condition: | Incorrect IP gets registered on the device and hence EFA doesnt receive syslog messages from the switches and hence delayed updates. | ||
| Workaround: | Users should ensure that the HOST IP to be registered with the devices as syslog server should be the 1st in the list when executing hostname -I on the EFA server. | ||
| Recovery: | Delete the syslog server that was configured and using efa `execute cli` configure the right IP to be use as the IP for syslog server on the devices. | ||
| Parent Defect ID: | EFA-3743 | Issue ID: | EFA-3743 |
| Reason Code: | Working as Designed | Severity: | S2 - High |
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.1.0 |
| Symptom: | Config Gen reason CLI does not show the detailed reasons if device is in refresshed state | ||
| Workaround: | EFA fabric show CLI shows these reasons, user can refer to that instead of the debug config-gen reason CLI | ||
| Parent Defect ID: | EFA-5830 | Issue ID: | EFA-5830 |
| Reason Code: | Not Reproducible | Severity: | S2 - High |
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | Unable to update devices using command : inventory device update --fabric {fabric name} | ||
| Condition: | When device update command is submitted immediately after fabric is created, intermittently it returns error with message "fabric" does not exist | ||
| Workaround: |
Use the device update command by specifying the ip instead of fabric name e.g. efa inventory device update --ip {comma separated list of ip addresses} |
||
| Recovery: | This issue is intermittent and resolves in few minutes. If it persists, delete the devices from fabric followed by deleting the fabric. Wait for 2 minutes. Recreate the fabric. | ||
| Parent Defect ID: | EFA-5859 | Issue ID: | EFA-5859 |
| Reason Code: | Already Reported | Severity: | S2 - High |
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | "efa tenant epg show" CLI execution time increases on each session with the number of parallel executions. | ||
| Condition: | Execute "efa tenant epg show" CLI in parallel from multiple sessions. | ||
| Workaround: | Do not execute "efa tenant epg show" CLI in parallel from multiple sessions. The CLI executions will be queued and processed sequentially. | ||
| Parent Defect ID: | EFA-5941 | Issue ID: | EFA-5941 |
| Reason Code: | Already Reported | Severity: | S2 - High |
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | If large number of Endpoint Groups are created and many of them use same VRF, the 'tenant epg show' output displays EPGs in 'cfg-refresh-err' state for some time after Endpoint Group creation. The state moves to 'cfg-in-sync' few minutes after all Endpoing Groups are created. | ||
| Condition: | This issue is seen when multiple Endpoint Groups are created and commands are run one after the other. The time taken to move Endpoint Group status to cfg-in-sync is impacted by number of EFA config and show commands that are being run. | ||
| Workaround: | Workaround is to wait for about 30 minutes after all EFA config and show commands have executed. | ||
| Recovery: | No recovery required | ||
| Parent Defect ID: | EFA-5949 | Issue ID: | EFA-5949 |
| Reason Code: | Already Reported | Severity: | S2 - High |
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | "efa tenant epg show" CLI remains hung for a long time. | ||
| Condition: | Execute "efa tenant epg show" CLI for a tenant having more than 100 EPGs | ||
| Workaround: | Execute "efa tenant epg show" CLI for a given EPG instead of executing it for a given tenant. | ||
| Parent Defect ID: | EFA-5954 | Issue ID: | EFA-5954 |
| Reason Code: | Insufficient Information | Severity: | S3 - Medium |
| Product: | Extreme Fabric Automation | Reported in Release: | EFA 2.2.0 |
| Symptom: | firmware-download prepare add/remove command fails with error message "Failed to find fabric with name: {fabric name}" even when fabric with this name exists | ||
| Condition: | When firmware-download prepare add/remove command with fabric name as parameter, is submitted immediately after the fabric is created, intermittently it returns error with message "fabric" does not exist | ||
| Workaround: |
Use the firmware-download prepare command by specifying the ip instead of fabric name e.g. efa inventory device firmware-download prepare remove --ip {comma separated list of ip addresses} |
||
| Recovery: | This issue is intermittent and resolves in few minutes. If it persists, delete the devices from fabric followed by deleting the fabric. Wait for 2 minutes. Recreate the fabric. | ||