OPNsense Forum
Archive => 19.1 Legacy Series => Topic started by: logo2002 on January 31, 2019, 10:56:57 pm
-
Hello guys,
I did the upgrade to the latest version 19.1 on my Opnsense firewall at home but since that upgrade I noticed that I got lot of ERR_CONNECTION_RESET in my web browser.
I thought it was an issue with the website/DNS or other but Ping command to these sites are OK and the DNS server is working fine to (8.8.8.8).
Do you have any idea of what can be the root cause of this issue since the configuration stays the same ?
Thank you,
Logan
-
I have the same exact problem
-
Same problem here, this is what i tried, unfortunately had to revert back to the previous release. I just did an upgrade through the gui, no clean install
This is my config
- bridged lan interfaces
- opendns nameservers configured
- bind for blocklist
- unbound in forwarding mode.
This did not work from lan disconnects and slowness all around. I reverted to the unmasq dns but this also did not help. Changed dns servers to google dns but also did not help. Did revert to my backup of the previous release and all is working well
If wanted i can share my config file with the dev's
-
Same issue here
Also I see blocked localhost traffic in the firewall log, that i never saw before the upgrade. And some blocked VLAN traffic
I can share logs if needed
-
scrensen and tiki1980, please open new threads as they are different.
logo2002 do you use proxy? Does telnet to the fqdn with port 80 or 443 work?
-
Hi Mimugmail,
I am not using any proxy and I already check if some of them were activated after the upgrade.
Ping to the hostname/IP is ok.
Nslookup of the hostname is ok.
Telnet to port 443 of the host is ok.
It seems that something in the firewall is stopping the connection to force the connection state to stop/reset.
I tried to restore the backup done before the upgrade when everything was working well but it doesn't change anything to my issue.
Nothing is blocked by the firewall rules (checked in the logs).
Thank you for your help.
Regards,
Logan
-
No I have the same issue
Strangely enough a curl from my MacBook commandline works just fine. But via three different browsers I have this issue.
And literally happened directly after the upgrade. Before that all was fine.
And when I connect my laptop directly to my cable modem it also works perfectly fine. I can not think of any other reason then the 19.1 upgrade
I made some traffic captures and at first glance I see a tcp window update and after that many TCP retransmissions.
Where as the successful session via curl just shows the ‘normal behaviour’
-
No I have the same issue
Strangely enough a curl from my MacBook commandline works just fine. But via three different browsers I have this issue.
And literally happened directly after the upgrade. Before that all was fine.
And when I connect my laptop directly to my cable modem it also works perfectly fine. I can not think of any other reason then the 19.1 upgrade
I made some traffic captures and at first glance I see a tcp window update and after that many TCP retransmissions.
Where as the successful session via curl just shows the ‘normal behaviour’
Interfaces : LAN : MSS
Set to 1400 and test again ...
Shouldn't be the issue, just curious
-
Thanks for the response
Didn't change anything.
Another thing to make it even more strange. I tried google.nl like 5 times in Safari and no go. While curl works just fine.
The tried Firefox again and it work for about three times. And then it stopped working. Same error.
As said, DNS works fine all the time. Could it have to do with some TCP buffers or something similar?
-
Same here, and I can't rollback because it says "failed, no signature found"
Someone know how to do it?
-
@All: experienced the same. Clear Chrome browsing history (assuming you use chrome) solved my issue.
https://www.webnots.com/how-to-fix-err_connection_reset-error-in-google-chrome/
-
@All: experienced the same. Clear Chrome browsing history (assuming you use chrome) solved my issue.
https://www.webnots.com/how-to-fix-err_connection_reset-error-in-google-chrome/
Thx!
-
@Northguy
I have this issue on all of my devices.
- Google Home
- Windows PC
- MAC laptop
- Playstation 4
Some website are available and some not.
-
Which are not? Only HTTPS?
-
Facebook is not working but youtube is.
-
My Iphone is unable to reach the apple store but can access some website.
I'm sure that I'm not the only one getting this issue
-
Do you have Realtek NIC?
Can you check this:
https://forum.opnsense.org/index.php?topic=11425.0
-
Yes and yes the MTU is set to 576 on my WAN.
I tried to force the MTU to 1500 on the interface but in CLI with ifconfig the interface stays at 576 even if I reboot thé opnsense.
I saw on internet that this settings are transmitted by DHCP on WAN are they any ways to disable the 576 ?
Thanks
-
You should be able to set it via UI in Interface config
-
Yes that’s what I do but after that I am loosing packet to internet (Ping command).
And I check with ifconfig and the interface stays at 576 even if it’s configured on GUI
-
Issue coming from MTU set to 576 by DHCP on WAN.
See post : https://forum.opnsense.org/index.php?topic=11425.0
-
In Interface Config under DHCP, Client Config, Advanced, set:
supersede interface-mtu 0
in Option Modifiers ...
Then try again ..
-
@Mimugmail
Thank you so much you solve my issue !
With this option the MTU keeps running at 1500 using the DHCP.
Thank you again
-
@All: experienced the same. Clear Chrome browsing history (assuming you use chrome) solved my issue.
https://www.webnots.com/how-to-fix-err_connection_reset-error-in-google-chrome/
Thx!
It was a temporary fix and ended up in the same issue as all the others. MTU 576 on WAN. WAN configured as DHCP on APU2D4 (intel NICs).
Applied supersede interface-mtu 0 in Option modifiers of WAN interface, which fixed the "problem"
-
After aplying "supersede interface-mtu 0" and rebooting, I get continuosly this kind of error:
Configuring WAN interface...ahcich0: Timeout on slot 6 port 0
ahcich0: is 000000 cs 0000000 ss 00000040 rs 00000040 tfd 50 serr 00000000 cmd 0000d717
(ada0:ahcich0:0:0:0): SEND_FPDMA_QUEUED DATA SET MANAGEMENT. ACB: 64 01 00 00 00 40 00 00 00 00
(ada0:ahcich0:0:0:0): CAM status: Command timeout
(ada0:ahcich0:0:0:0): Retrying command
and my SSD keeps R/W all the time, and I lost all connectivity, even LAN
For the record, It´s a clean install
-
That sounds like a BIOS/coreboot issue, what hardware are you running it on ?
-
I´m using this:
http://www.xcyminipc.com/product/showproduct.php?lang=en&id=51
Had no problems with 18.1 and 18.7
It all started with 19.1 (upgraded or clean install)
-
1. Check for a new BIOS
2. See if the mitigation steps here apply to your CPU as well
https://forum.opnsense.org/index.php?topic=11419.msg52375#msg52375