I've upgraded my MX85 from 17.9 to 17.10. Unfortunately I observed that after the upgrade to 17.10, my uplink WAN connection (1x PPPoE to a single ISP) has capped to around 100 Mbps download and 100 Mbps upload. With 17.9, I manage to get the true speed subscribed from ISP which is 800 Mbps download and 200 Mbps upload.
Now I fallback my MX85 to 17.9 and everything back to normal speed.
@incinrea that is interesting, on the MX75 I have a 500/500 line and it is still running at 500 both ways. No PPPoE though...
right. I noticed some user that downloaded MX18.101 beta has experience similar experience where the WAN speed has been capped into certain limit.
Unfortunately, the content has been removed.
FYI, I've PPPoE with IPv6 capabilities.
hi, I just want to update you and the team here that I've tried to upgrade my MX85 from 17.9 to 17.10 again and this time I want to report that my I manage to get the true PPPoE connection speed, i.e. Download 800 Mbps and Upload 200 Mbps.
So far so good.
I wonder when this will get fixed.
- Client traffic will be dropped by MX65(W), MX67(C,W), and MX68(W,CW) appliances if 1) The client is connected to a LAN port with 802.1X authentication enabled and 2) The VLAN ID of the port is configured to 16, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, or 240.
Fortunately this has been fixed from 17.9:
We'll wait for 17.10 to become the stable version.
MX67W updated to 17.10 but wireless clients experiencing extreme slowness. Had to roll back to 17.9.
See also this community thread: After firmware update 17.10 - WLAN, iOS devices are slow and App Store runs... - The Meraki Communit...
I think Im seeing some strange behaviour on MX84 and Layer 7 Firewall rules.
The layer 7 firewall seems to block internal DNS requests (Clients on one vlan - dns server on another). DNS seems to be classified as eDonkey P2P traffic ... 😕
To external DNS servers it seems to work fine ... very strange.
Im also seeing some strange behaviour on MX85 where it will stop traffic for small periods of time during the day, seems to be related to IPS rules that is being hit a lot of times all of a sudden, but Im more unsure of this one, --- It could also be a coincidence , and its in reality the WAN switch in this setup that is running out of buffer (for some reason).
Hmmmm ... I think 17.10.2 reintroduced the EEE bug on the MX85 boxes I monitor ...
From just around the update I have started to see a lot of "Ethernet port carrier change" in the eventlog.
I will try to "investigate" some more ....
You need to check Content Filtering for the sites. As with upgrade Meraki is pushing some additional Content filtering groups on the site which impact Internet and IPSEC traffic (Syspro & Flowcentric ).
Removing Content filtering categories resolved the issue.
Also check Layer 7 rule and change Peer to peer traffic
All our MX's (64, 65, 67, 68, 84, 100) have been running 16.16.7 for a while. I have tried to upgrade to 17.10 on several customers, but they all encountered slow dns performance. Basically because i could not change the content filtering setting to Top sites only (better performance). I had to roll-back the upgrade to 16.16.7. Meraki support told me in the next version it will be solved. Does anyone know if the slow dns/category filtering problem is solved in 17.10.2 ?
Migrated 2 customers to 17.10.2 and the first test and production weeks reveals no dns performance problems.
I had to adapt/change/add all the "content filtering" categories as they were change to the Talos cloud solution.
Apart from that, its looking good.
It seems that there is a problem with the 17.10.x Version when a iOS Devices uses WIFI to connect to the INet. The performance is incredible slow.. Is there a chance that this will getting fixed soon?