Not on topic, but I think a new Techincal forum should be created just for Access Manager. ... View more

So basically the cure was worse then the disease 🙂   ... View more

The switch was apparently not happy with my IPv6 config. "Router FE80::AA46:9DFF:FEB6:2137 on Vlan10 conflicting ND setting prefix somethinsomething::/64 valid lifetime, difference 2586600 seconds" I have no idea what this means 🙂 ... View more

PPS: Im also super happy that the switch now apparently have a whole bunch of static IPv6 routes to China. (there are also some other ones that I can lookup to Meraki ... fx. 2606:7BC0::/32 hot dangit thats a lot of IPs).   S 2402:1440:40:100::/64 [1/0], tag 2896997548 via FE80::AA46:9DFF:FEB6:2137, Vlan10 S 2402:1440:40:115::/64 [1/0], tag 2896997548 via FE80::AA46:9DFF:FEB6:2137, Vlan10 S 2402:1440:40:531::/64 [1/0], tag 2896997548 via FE80::AA46:9DFF:FEB6:2137, Vlan10 S 2408:8706:0:1281::/64 [1/0], tag 2896997548 via FE80::AA46:9DFF:FEB6:2137, Vlan10 S 2408:8706:0:1282::/64 [1/0], tag 2896997548 via FE80::AA46:9DFF:FEB6:2137, Vlan10   Im GUESSING that these are the IPs used in Merakis Chinese DC ? But I do not know 🙂   ... View more

Hmmmm the switch might have had some problems (speculation at this point) with the static IPv4 address it had before the upgrade. After the upgrade (and I only looked at it now) it had a strange 198.x.x.x IP as "LAN" (not anything we have configured anywhere), and it had a IPv6 (we run dual stack) and it was using the IPv6 address. Speculation : could this be caused by the new way that static IPs are configured on a Catalyst switch (now under DHCP and routing ?), perhaps ? I have "re-saved" the config that was on that page, and now the switch has its IPv4 address just fine. (The IPv6 one did now not work , for some reason now, but Ill play around with it some more 🙂 ) ... View more

So ... dont upgrade unless you have many network flows because the fix for that was bad ??? Im a little confused here 🙂 ... View more

PS: Since the device uptime is now around 7 hours, the dashboard outage from 5:15 to 5:18 was not due to the switch reloading ... I have no idea what happened there. Could in theory have been something else. ... View more

I can hardly wait to upgrade the next stack ... a real thrill ride 🙂 ... View more

Oh ..-- and also this line on "top". Mar 19 03:17:17 sw01.lyst [2] FMFP 3 OBJ_DWNLD_TO_DP_STUCK Switch 1 F0/0: fman_fp_image: AOM download to Data Plane is stuck for more than 1800 seconds due to resolve object: obj[304] type[32] 'Void Intf (6)', resulting in it being a pending-issue object   For the outage at around 5:15 to 5:18 CET there is no log .. so who knows what happened there 🙂   From my very quick look at the logs, it failed the upgrade, and the 7200 second auto abort kicked in. Then it did a new install, some "DATABASE_MEMLEAK" happened, but it completed, both switches got back and the found out they where in a stack, and it did "commit". - So "success".... I guess 🙂 But what happened at 5:15 where it was offline, I cannot tell you from the logs, because there is no log from thie period of time. ... View more

So from the log ... At around 22:30 , this is where i tell it to schedule the upgrade at 23:00, start upgrading and finish the upgrade.   I have just collected the complete thing here, and omitted irrelevant things from the log, I have not analyzed it yet   Mar 19 02:52:24 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Dashboard Upgrade Final Status - upgraded Mar 19 02:52:24 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Image Install Commit Completed <some output omitted> Mar 19 02:51:29 sw01.lyst [2] INSTALL 5 INSTALL_COMPLETED_INFO Switch 2 R0/0: install_mgr: Completed install commit Mar 19 02:51:25 sw01.lyst [2] NDBMAN 5 MODEL_RPC_I Switch 2 R0/0: ndbmand: RPC 'http://cisco.com/ns/yang/Cisco-IOS-XE-install-rpc:install-commit' invoked from fd0a:9b09:1f7:1:248e:d42e:f6b6:9ae1:35190 via NETCONF user meraki-user session 28 <some output omitted> Mar 19 02:51:24 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Starting Image Install Commit <some output omitted> Mar 19 02:49:48 sw01.lyst [2] RF 5 RF_TERMINAL_STATE Terminal state reached for (SSO) Mar 19 02:49:47 sw01.lyst [2] HA_CONFIG_SYNC 6 BULK_CFGSYNC_SUCCEED Bulk Sync succeeded Mar 19 02:49:44 sw01.lyst [2] PKI 6 AUTHORITATIVE_CLOCK The system clock has been set. <some output omitted> Mar 19 02:49:27 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Image Install Activate Completed <some output omitted> Mar 19 02:48:37 sw01.lyst [2] REDUNDANCY 5 PEER_MONITOR_EVENT Active detected a standby insertion (raw-event=PEER_REDUNDANCY_STATE_CHANGE(5)) Mar 19 02:48:36 sw01.lyst [2] REDUNDANCY 5 PEER_MONITOR_EVENT Active detected a standby insertion (raw-event=PEER_FOUND(4)) Mar 19 02:48:27 sw01.lyst [2] STACKMGR 6 STANDBY_ELECTED Switch 2 R0/0: stack_mgr: Switch 1 has been elected STANDBY. Mar 19 02:48:26 sw01.lyst [2] IOSXE_REDUNDANCY 6 PEER Active detected switch 1 as standby. <some output omitted> Mar 19 02:48:26 sw01.lyst [2] IOSXE_REDUNDANCY 6 PEER Active detected switch 1 as standby. <some output omitted> Mar 19 02:43:15 sw01.lyst [2] INSTALL 5 INSTALL_COMPLETED_INFO Switch 2 R0/0: install_mgr: Completed install activate <some output omitted> Mar 19 02:40:08 sw01.lyst [2] INSTALL 5 INSTALL_AUTO_ABORT_TIMER_PROG* Switch 2 R0/0: install_mgr: Install auto abort timer will expire in 7200 seconds Mar 19 02:39:48 sw01.lyst [2] INSTALL 5 INSTALL_AUTO_ABORT_TIMER_PROG* Switch 1 R0/0: install_mgr: Install auto abort timer will expire in 7200 seconds Mar 19 02:39:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: C2E89FA4402A8017 Mar 19 02:37:33 sw01.lyst [2] NDBMAN 5 MODEL_RPC_I Switch 2 R0/0: ndbmand: RPC 'http://cisco.com/ns/yang/Cisco-IOS-XE-install-rpc:activate' invoked from fd0a:9b09:1f7:1:ceb:c403:d49b:d786:60746 via NETCONF user meraki-user session 34 Mar 19 02:37:32 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Starting Image Install Activate <some output omitted> Mar 19 02:34:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: C2E89FA4402A8017 Mar 19 02:29:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: C2E89FA4402A8017 Mar 19 02:24:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: 811459B41BAF0013 Mar 19 02:19:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: 811459B41BAF0013 Mar 19 02:14:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: 811459B41BAF0013 Mar 19 02:10:37 sw01.lyst [2] CRYPTO_ENGINE 4 CSDL_COMPLIANCE_RSA_WEAK_KEYS RSA keypair CISCO_IDEVID_CMCA_SUDI is in violation of Cisco security compliance guidelines and will be rejected. Mar 19 02:09:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: 811459B41BAF0013 <some output omitted> Mar 19 02:04:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: 811459B41BAF0013 Mar 19 01:59:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: 811459B41BAF0013 Mar 19 01:54:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: 811459B41BAF0013 Mar 19 01:49:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: 811459B41BAF0013 Mar 19 01:45:25 sw01.lyst [2] INSTALL 5 INSTALL_COMPLETED_INFO Switch 2 R0/0: install_mgr: Completed install add flash:/switch-iosxe-c9k-V1715_2M1_FC5 Mar 19 01:45:25 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Image Install Add Completed <some output omitted> Mar 19 01:44:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: 811459B41BAF0013 Mar 19 01:42:26 sw01.lyst [2] NDBMAN 5 MODEL_RPC_I Switch 2 R0/0: ndbmand: RPC 'http://cisco.com/ns/yang/Cisco-IOS-XE-install-rpc:install' invoked from fd0a:9b09:1f7:1:4217:6160:e475:bc3d:44004 via NETCONF user meraki-user session 31 <some output omitted> Mar 19 01:42:26 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Starting Image Install Add Mar 19 01:42:26 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Image Download Completed <some output omitted> Mar 19 01:40:46 sw01.lyst [2] FMFP 3 OBJ_DWNLD_TO_DP_STUCK Switch 1 F0/0: fman_fp_image: AOM download to Data Plane is stuck for more than 1800 seconds due to resolve object: obj[293] type[32] 'Void Intf (6)', resulting in it being a pending-issue object Mar 19 01:40:44 sw01.lyst [2] FMFP 3 OBJ_DWNLD_TO_DP_STUCK Switch 2 F0/0: fman_fp_image: AOM download to Data Plane is stuck for more than 1800 seconds due to resolve object: obj[310] type[32] 'Void Intf (6)', resulting in it being a pending-issue object Mar 19 01:39:27 sw01.lyst [2] CRIMSON 3 DATABASE_MEMLEAK Database memory leak detected in /tmp/rp/tdldb/0/IOS_PRIV_OPER_DB database, Basic type/No extension/N/A size:32 byte, callsite: C2E89FA4402A8017 <some output omitted> Mar 19 01:36:33 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Starting Image Download Mar 19 01:36:31 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Image Install Clean Completed <some output omitted> Mar 19 01:35:33 sw01.lyst [2] INSTALL 5 INSTALL_COMPLETED_INFO Switch 2 R0/0: install_mgr: Completed install remove Mar 19 01:35:33 sw01.lyst [2] NDBMAN 5 MODEL_RPC_I Switch 2 R0/0: ndbmand: RPC 'http://cisco.com/ns/yang/Cisco-IOS-XE-install-rpc:remove' invoked from fd0a:9b09:1f7:1:4854:582e:3de2:e6c4:33638 via NETCONF user meraki-user session 28 Mar 19 01:35:32 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Starting Image Install Clean <some output omitted> Mar 19 01:35:31 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Dashboard is initiating Image Upgrade version_name-switch-iosxe-c9k-V1715_2M1_FC5 Mar 19 01:34:08 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Dashboard received an Image Upgrade Request version_name-switch-iosxe-c9k-V1715_2M1_FC5 <some output omitted> Mar 19 01:34:05 sw01.lyst [2] PKI 6 AUTHORITATIVE_CLOCK The system clock has been set. <some output omitted> Mar 18 23:32:35 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Dashboard Upgrade Final Status - unreachable Mar 18 23:32:35 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Image Install failed - Device is unreachable Mar 18 23:05:47 sw01.lyst [2] INSTALL 5 INSTALL_COMPLETED_INFO Switch 2 R0/0: install_mgr: Completed install activate <some output omitted> Mar 18 23:02:39 sw01.lyst [2] INSTALL 5 INSTALL_AUTO_ABORT_TIMER_PROG* Switch 2 R0/0: install_mgr: Install auto abort timer will expire in 7200 seconds Mar 18 23:02:17 sw01.lyst [2] INSTALL 5 INSTALL_AUTO_ABORT_TIMER_PROG* Switch 1 R0/0: install_mgr: Install auto abort timer will expire in 7200 seconds <some output omitted Mar 18 23:00:02 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Starting Image Install Activate <some output omitted Mar 18 22:43:18 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Image Install Add Completed <some output omitted> Mar 18 22:37:24 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Starting Image Install Add Mar 18 22:37:23 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Image Download Completed <some output omitted> Mar 18 22:31:31 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Starting Image Download Mar 18 22:31:28 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Image Install Clean Completed <some output omitted> Mar 18 22:30:32 sw01.lyst [2] INSTALL 5 INSTALL_COMPLETED_INFO Switch 2 R0/0: install_mgr: Completed install remove <some output omitted> Mar 18 22:30:31 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Starting Image Install Clean <some output omitted> Mar 18 22:30:29 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Dashboard is initiating Image Upgrade version_name-switch_combo_4311_4429 Mar 18 22:29:44 sw01.lyst [2] DASHBOARD_IMG_MGMT 5 INSTALL Dashboard received an Image Upgrade Request version_name-switch_combo_4311_4429   ... View more

I just did some 17.15.1m1 (the EFT release of native NexTunnel) to this official release with some ehhh mixed results.   A 9200L stand alone, no problems, upgraded as expected, nothing to see here 🙂   A 9300L stack uh oh .. this was a little "wonky". The Switch seemed to upgrade just fine, and the equipment connected to the switch (APs, cameras, and clients) came online just fine and the whole thing was down for about perhaps 6 minutes. BUT ... the switch itself never got back online on the dashboard. So this was kinda strange to me, because as far as I know the new UAC feature and so on should "just work(tm)".   With everything working, except the switch not being online in the dashboard, i created a case, and went to bed. This morning the switch IS online again, and the uptime graph looks like this.   Thats kinda strange right.   So of course there is some differences here versus the stand alone C9200L. This was a C9300L stack, and this stack was setup with a static IP instead of DHCP, but that should of course not matter.   Im currently looking at logs to see if I can figure out what happened 🙂 ... View more

Regardless of the answer, have anyone ever done it using a couple of fx. Multimode 40g qsfp optics ? ... View more

I know, unfair to accept your own post as the solution. But until I know more, this was the easiest way to show others my experience. ... View more

Well cant get the MX to ping Umbrella from its VLAN interfaces. But I can get an AP on the management VLAN to ping Umbrella IPv6 ... so I guess everything works. ... View more

Set the static IPv6 on the VLAN interface. After a while they where also changed on the IPv6 prefix VLAN assignment page. VLAN interfaces still have a link local and multicast address. Should be fine ?   I dont quite know if the MX supports IPv6 pinging from a VLAN interface ... View more

My second site. Same provider. Is also broken on IPv6. No VLAN assignments on the prefix page, even though the IP subnet has been added on each vlan. All VLANs do not have the link local or multicast address.   Removed IPv6 network from all VLANs. - or rather I set it to auto. Removed the prefix from the IPv6 prefix page. Set the WAN connection to auto for fun. Set it back to Static Added prefix again on the prefix page.   Waited to see what would happen. After a while, all VLANs now have an auto assigned /64 from the /48 prefix seems fine.   I will now try to set the vlan to static IPv6 to see if I can get this to break again, and start over. There seems to be a "order of operation" that breaks something somewhere. ... View more

Even the vlan page looks more "right" now. With Link local and Solicited-node multicast address (they where not there before, only the manual IPv6 address I assigned).   What happened ? reboots and patience ? I have no idea ... View more

This is what the IPv6 Prefix page looks like now, after waiting and reboots. I have NO idea where the Reserved prefixes came from. ... View more

Well i add a prefix, a /48 for WAN1, on the IPv6 Prefix page of appliance status. Then I go add a /64 from that /48 on each vlan setting it as manual WAN1 on the vlan page. And then it behaves strangely.     ... View more

Hmmm perhaps my ISP does do PD ..... I was told they did not. Stand by 🙂 ... View more

After applying the correct IPv6 addresses over and over to one of the VLANs it finally showed up correctly in the VLAN assignments on the prefix page, and that vlan now also has link local address and so on. But the other vlans, not so much. The fun thing, I can still do ping -6 from these vlans, and I get the correct IPv6 address as I expect. But for how long .... time will tell. ... View more

You can have STP blocking one of the links to your switch. If you configure the same settings on the two links on the MX, and have spanning-tree enabled on your switch (it should be by default), then your switch should block one of the links. In case of the active link fails, the other link should then be opened. ... View more

Oh .. PS: and because you have a free try to do a Cisco test, I tried Engineering Cisco Meraki Solutions (ECMS 500-220) and passed 🙂 Weeeeee !!!! 🙂 ... View more