I have an ongoing Allscan "ASL stats HTTP Error: Failed to connect to stats.allstarlink.org port 80 after 40 ms: Couldn't connect to server. Will retry in 15 Seconds..." on all 5 of my ASL3 nodes.
Allscan will scan the favorites & update the stats intermittently off and on. At times it will scan for a few hours, but most of the time it will fail to connect to the ASL stats server frequently and cease to update the list of favorites for an inconsistent interval.
Currently, I am only running 1 instance of Allscan, the other 4 nodes are using Allmon3 for the time being.
There are only 15 favorites in my Allscan favorites list. It appears that the dynamic request timing algorithm within Allscan is not running correctly to limit the ASL stats API to 30 requests/minute per IP Address. So, the ASL stats server will temporarily block my stats requests for an interval of time.
I have completely removed Allscan, and re-installed 2 times already.
Since Allscan will work correctly for periods of time, I believe that there is no issue with my outgoing port 80 or port protocol on my router, or within the ASL3 firewall. Additionally, I believe my ASL3/Allscan config files are correct. Also, my fiber internet connectivity is rock solid.
I do allow friends that are on a different public IP address, access to my hub dashboard. But, I'm thinking that their long list of favorites has no bearing on my node limit of 30 requests/minute per IP Address. Besides that, the algorithm should circumvent that.
Obviously, I do have a problem; and I need some help.
This has been a common issue as the ASL stats server(s) have been upgraded over time, I don't know the details of why or when, but I can say that the issue is not with the AllScan web app. You can make the same http request directly in a browser or eg. with wget and you'll see the same thing ie. server not responding. The ASL devs might know more info on why this is happening. I do know that some nice updates are now in process - one of which will allow stats for all nodes in the Favs table to be returned in one request rather than having to do individual requests. This will allow the stats info for the whole table to update almost instantly instead of taking several seconds per row, which will be a very nice enhancement. There is another thread on the forum with details on the stats server API upgrades.
The other thing you mentioned was the stats API request rate limiting algorithm in AllScan. This works great if there are only 1 or 2 web users using an instance of AllScan ie. that is running on one server with a specific IP address. If you have 3+ users all viewing the same instance at the same time that will definitely slow down the stats requests as the 30 requests per minute rate limit will be exceeded much more quickly. Once the aforementioned stats APIs enhancements are completed and AllScan updated to use those this should no longer be an issue. Also, note that AllScan is intended mainly for use on personal nodes by one user, in contrast to other dashboards such as Allmon and Supermon which are intended more for managing multiple nodes / hubs / larger systems. AllScan thus does not do any stats API data caching that would be needed to better support multiple simultaneous users, and that is outside the scope of AllScan's intended use cases, at least for now.
Thank you very much for your information. I will search and follow the thread(s) relative to this subject. The <5 users that I allow access to see my dashboard are all on separate public ip addresses, so I didn't think that would be a problem.
I have been fighting this issue for quite sometime, and frankly I was at wits end! I will be patiently awaiting ASL developers to address this issue. Thanks, and 73 Sir.
How many AllScan browser windows do you have open across all of the desktops, laptops, and mobile devices on your network? I'm asking because here's a 1 minute snapshot of the queries your network is making for a single node :
When you add in the other nodes being queried, multiple times per minute, your IP ends up being tagged as making too many requests and gets blocked for an interval of time.
I have been trying to only have 1 instance of Allscan running at a time on my network. I just realized that I had 2 other instances running on my 2 ipads after reading your post, for a total of 3. I now only have 1 instance running now. Will you please run that query on the server again?
I wish I had the capability to fetch and know how to read and understand what you queried the server. That would be a great tool for me to use.
This looks like 156.146.218.145 is requesting api stats for node 40131 on 23/sep/2026. I was monitoring the 40131 dashboard a couple of days ago, but not yesterday or today at all. What is the date and time that this is for?
The logs that I captured included the [UTC] date/time of those requests. Looking at the recent logs I still see your IP address making many requests for the same nodes in a short period of time.
How many browser windows do you have running AllScan? and don't forget about those that are minimized, in tabs that you are not viewing, etc.
Do you have any related "apps" running that might be querying the status of AllStarLink nodes, nets, ...?
I have checked all of my computers, phones, & ipads, and I am only seeing 1 computer with 1 instance of Allscan running. And, I have now reduced my favorites list from 15 nodes to 4 nodes.
As far as related apps, I do use the Allmon3 dashboard on my 4 other nodes. But, using Allmon3 on my other 4 nodes concurrently with Allscan, should not have any adverse effects with my Allscan issue, right?
Right now, I only have my 607811 & 607812 node online. 607812 is not currently using any dashboard. It just provides rf access to the radioless hub 607811.
On a node that I temporarily have offline due to testing with this issue, I do stream that audio to broadcastify. But, it is offline, and I wouldn't think that would impact my issue when it is active.
Finally, on one of my ipads, I have saved the 5 nodes to my home screen as what I think is only a shortcut link. When I swipe up on the bottom of the ipad to use the app switcher, any of those 5 do not show up as being active. If you think that may be an issue, I will delete those.
Once again, thank you very much with your time, patience, and expertise.
The IP address is pretty aggressively pulling down stats data. Right now, in the last 60 seconds that IP address asked for something from stats 33 times or once every 2 seconds.