I remotely upgraded one from 6.9.5 to 6.10 this week and lost remote access.
The dynamic DNS address is working correctly for all other services
When I had physical access I did the three button presses for a shutdown, when it rebooted it came back up.
I tested remote access and thought it was working.... but it isn't now
At some point there has been a feature enabled to allow automatic hotfix updates, enabled by default. I switched mine off as I want to know if a software update has screwed things up.
The reason I noticed was because it already said "Version 6.10 Hotfix 2"
Looks like a poor version release, not out long and already on a second hotfix.
Netgear seem to be having problems in this regard lately
Re-reading the release notes it seems I cannot downgrade, what are my options.
I am not an early adopter as I like bugs to be fixed before I deploy "upgrades"
I will not be upgrading my RN104 which is still on 6.9.4 following problems some people had with 6.9.5.
I currently have no access to the RN102 remotely, it's about 10 miles away from me.
Reading around it seems 6.10 has severe problems - I wish I had searched more dilligently before upgrading, and given these problems I suggest 6.10 is pulled until the problems are fixed.
I think most upgrade issue with the RN100 series are down to a full OS, which mostly is caused by a bug with AV software - later hotfixed.
Anyhow, as you have no access to the NAS it is hard to do anything for you 😞
I reckon you need to go and give it a simple reboot to start with and see where you land. Any chance you can "bring it home" for better troubleshooting?
Sounds like I need to examine the size of the system drive then.
I am unable to access the NAS until Monday, so I will see if it is working for local connections first (which I expect will be fine) and then reboot and test from an internet connection again.
Which particular log is of interest. I can extract those and post them up here on Monday morning with a bit of luck
Just as an FYI I am port forwarding to the NAS using a port in the 50000-60000 range - this has been working without issue before, but I know this is not the default setup
Posting from my second account as I can't remember / have stored the password for my main account on this PC
I have rebooted the NAS and performance is as before.
Used SSH access to run a df -h command, this shows /dev/md0 has 644MB used of 3.7GB available so only 19% used
Next I checked /var/lib/clamav for AV definitions. None there completely empty which confirms I have never used it
Dynamic DNS is functioning correctly and by way of confirmation, other apps / services can be connected as normal, it's just the Readynas remote admin/shares which are no longer working
Are there any particular logs I should upload - I am with the NAS now
I have sent the message as suggested, having downloaded the full logs through the admin console - it's only remote admin that is not working
I'm confused on what you mean by "admin console" and "remote admin". When you say "admin console", do you mean that the admin web ui ui works locally? Or do you mean something else?
Apologies if I am using confusing terminology, allow me to explain. (and back on the other PC/profile)
This NAS is not at home it's at a different location.
I have set up my router at the remote location with port forwards to allow access to the Admin Page and thus the shares when I am somewhere else. Rather than the default 443 I am using a port in the 50000-60000 range.
I did the upgrade remotely (which has worked before) and after a reboot I could not access the Admin page
When on the local LAN I can access the Admin Page, as before
Since the upgrade to 10.0.0 and a couple of reboots I am still unable to access the Admin Page over the internet (what I call remote access)
I am now at home and will duplicate the setup / port forwarding so I can test while at home, especially as I'm 5 to 8 hours ahead of the contiguous states time zones.
The RN102 is not working for remote access at home
I made configuration changes in HTTPS settings to my RN104 which is running 6.9.3, changed the LAN address the port forwarded to and tested the RN104
It worked , that's a relief
So, this is definitely an OS 6.10.0 issue......... based on browser responses my gut is telling me it's something to do with how SSL browser certificates are being handled ???
Personally I switched to openVPN years ago, so this isn't something I can easily test.
What settings are you using in the NAS and in the router? It should certainly work if you set the secondary port for HTTPS in the NAS, and forward that port in the router to the NAS. You will of course see the security warnings in the browser, but there is nothing new about that.
In the router I am forwarding traffic on TCP 55555 to the static IP address for the NAS
In the NAS I have set the second port to 55555, leaving the default 443 on port 1
At home it's a bit more complicated than that because I am double natted, one of the reasons I tested with the RN104.
So, I set up the RN104 (on 6.9.3) to use port 55555, and port forwarded to its static IP address.
Then I used a mobile phone, not on the network to attempt to connect using no-ip on xxxxxxx.ddns.net:55555
It connected to the RN104 with the usual security warnings. I accepted the security exception and was into the Admin Page.
Great, that confirms my network setup was correct.
Then retested the RN102 (6.10.0)
Changed the IP address for the port forward
Used the mobile phone to connect using the same URL
The browser did not prompt for a security exception and did nt complete connection.
Then I used the mobile phone as a hotspot for a laptop
Tested again in FF and a Chromium based browser
FF gave a "connection interrupted" and Chrome a different error message, I would need to repeat the test to get the exact message.
At this point the conclusion was clear, whereas in 6.9.5 the browser was getting an "insecure" connection, in 6.10.0 it just seems to be bombing out
Is this perhaps because of the change in Apache from 2.2 to 2.4 ?
I did some searching in the http logs, and found these bits which may be relevant, I have obfuscated IP addresses
This is from 6.9.5
Jan 26 16:19:39 Nas_102 apache2[2478]: apache2: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName
and this is from 6.10.0
Apr 30 12:35:38 Nas-102 systemd[1]: Started The Apache HTTP Server. Apr 30 12:35:38 Nas-102 apache2[2173]: [ssl:warn] [pid 2173] AH01909: ::1:443:0 server certificate does NOT include an ID which matches the server name Apr 30 12:36:02 Nas-102 apache2[2177]: [cgi:error] [pid 2177] [client 111.222.333.444:52684] AH01215: Connect rddclient failed: /frontview/lib/dbbroker.cgi, referer: http://111.222.333.444/admin/index.html
In non technical terms it seems that in 6.9.5 there was simply a warning about the certificate setup (I never changed from the existing setting in HTTPS, SSL Key Host), now in 6.10.0 there is an actual error in dbbroker.cgi
Does that help point the finger towards the fault or even a solution ?
I just posted a long reply and it seems to have disappeared.
If you can see it somwhere in moderation somewhere can you make it appear please. It isn't in drafts and when I posted it i got a pop up message for a new badge so I definitely posted it
OK Stephen, I can see you understand what I am doing 🙂
Let's assume that I am using port 55555, this is what I did
First RN104, running 6.9.3
In my router I set up a port forward for TCP 55555 to the static IP address for the NAS
In the NAS I went into https setup and set port 2 to 55555
Then I used a mobile phone, not on the network, to connect to the nas using no-ip on xxxxxxx.ddns.net:55555
I was prompted for a security exception, added the certificate / exception and successfully connected to the Admin Page
Good news, the network side is fine as I am actuallly double natted
Then I moved onto the RN102, running 6.10.0
Changed the IP address for the port forward and tried connecting with the phone
The phone browser failed to complete connection
Then I used the phone as a hotspot and tried with FF and a chromium browser.
FF failed to connect and chromium came up with an error message which i cannot remember, other than it suggested "try again"
At no point did I get the option to confirm a security exception
Could this be related to an Apache change from 2.2 to 2.4 ?
I have also been looking at the http logs
This is from the 6.9.5 log on the Rn102
Jan 26 16:19:39 Nas_102 apache2[2478]: apache2: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName Jan 26 16:19:39 Nas_102 apache2[2478]: Syntax OK Jan 26 16:19:39 Nas_102 systemd[1]: Reloaded The Apache HTTP Server. Jan 26 16:19:39 Nas_102 apache2[2318]: [warn] RSA server certificate CommonName (CN) `nas-35-90-D8.local' does NOT match server name!?
and this is from 6.10.0 - IP addresses obfuscated
Apr 30 12:35:38 Nas-102 systemd[1]: Started The Apache HTTP Server. Apr 30 12:35:38 Nas-102 apache2[2173]: [ssl:warn] [pid 2173] AH01909: ::1:443:0 server certificate does NOT include an ID which matches the server name Apr 30 12:36:02 Nas-102 apache2[2177]: [cgi:error] [pid 2177] [client 111.222.333.444:52684] AH01215: Connect rddclient failed: /frontview/lib/dbbroker.cgi, referer: http://111.222.333.444/admin/index.html
So is it perhaps something to do with the SSL Key Host in HTTPS settings, there was previous just a warning, now there is an actual error in dbbroker.cgi ????
Apr 30 12:35:38 Nas-102 apache2[2173]: [ssl:warn] [pid 2173] AH01909: ::1:443:0 server certificate does NOT include an ID which matches the server name Apr 30 12:36:02 Nas-102 apache2[2177]: [cgi:error] [pid 2177] [client 111.222.333.444:52684] AH01215: Connect rddclient failed: /frontview/lib/dbbroker.cgi, referer: http://111.222.333.444/admin/index.html
It does look like it might be a consequence of an uplift in security, though it seems odd that it would only show up remotely. Did you try a local connection using the secondary port?
You could also try regenerating the key by changing the name on the https settings page, and clicking "Apply". If the key host already matches the hostname, then try changing it back.
Setting up a PC on the NAS site that has some form of remote desktop installed (google remote desktop perhaps) might be worth doing, as it would give you another path to the NAS.
BTW, I was able to find your earlier post - it was caught by the automatic spam filter. I released it.
Many thanks for finding and releasing my "spam", I'm not sure what I did deserve that 🙂
Anyway, back to the original issue and your suggestion to do a simple LAN connection. A good idea by the way.
I tried connecting to the RN104 specifying port 443 and "55555". Both worked and I was prompted for the security exception each time
On the RN102 I was prompted as expected on port 443
I received the partially obfuscated error page below on FF when using "55555"
Unable to connect
Firefox can't establish a connection to the server at 192.168.xxx.yyy:55555.
The site could be temporarily unavailable or too busy. Try again in a few moments. If you are unable to load any pages, check your computer's network connection. If your computer or network is protected by a firewall or proxy, make sure that Firefox is permitted to access the Web.
This is the entry in the http log
May 01 16:51:46 Nas-102 apache_access[2412]: 192.168.xxx.zzz "GET / HTTP/1.1" 302
Isn't a 302 a moved /redirect http error code so the replacement URL should be provided to the requesting browser ?
Seems to be begining to look more like a failure of the second port to respond correctly than an SSL issue now ??
I received the partially obfuscated error page below on FF when using "55555"
Unable to connect
Firefox can't establish a connection to the server at 192.168.xxx.yyy:55555.
The site could be temporarily unavailable or too busy. Try again in a few moments. If you are unable to load any pages, check your computer's network connection. If your computer or network is protected by a firewall or proxy, make sure that Firefox is permitted to access the Web.
Seems to be begining to look more like a failure of the second port to respond correctly than an SSL issue now ??
Yes.
I've confirmed that the second https port won't connect on my RN526x also. I checked this with both Chrome and Firefox. Reconfiguring the second port doesn't change the behavior.
@JohnCM_S / @OOM-9: Can you log a bug for this (and hopefully provide a hotfix).
@937carrera: I suspect forwarding 55555 to port 443 of the ReadyNAS in your router will provide a work around.
I have concluded that my DGND3700v2 router cannot provide external port triggering combined with port forwarding, and as I don't want port 443 open to the world I am content to wait for the hotfix.
It's possible that the router at the remote location can do this so I may try that tomorrow.
While wandering around reading the logs I have spotted a new error occurring in the readynasd log, for example, this is the first occurrance and it arose in the reboot following upgrade to 6.10.0
Apr 25 23:16:10 Nas_102 readynasd[2673]: Failed to call method GetUnitFileState: No such file or directory
It occurs at random times from hours to seconds and also at regular 5 minute intervals.
Any ideas what this means - happy to create a new thread if that is preferred, or it may be related ???
Apr 25 23:16:10 Nas_102 readynasd[2673]: Failed to call method GetUnitFileState: No such file or directory
I don't know what this is either, perhaps @JohnCM_S or @OOM-9 will know.
Thanks for that.
I have reinstalled the RN102 in its usual location and set up the port triggering / forwarding as suggested as that router has additional capabilities. It worked.
This is a workaround, not a solution, so I would still hope that Netgear can fix this properly.
As I have not seen any response in the thread or received a PM what's the best approach for logging / following up on bugs?
, that's a relief
