That's very unfortunate as it's basically a requirement by a lot of pools for special interactions. Is it possible to do it through the UI instead of through the API as you used to be able to do as a fall back method? Or perhaps through SSH?

Hey Patrike, seems the bug regarding passing the " = " still has not been fixed for Braiins OS. If it's present in the pool it wont do anything. I have to remove it from the pool on AM, then manually go through every miner and readd that to them if it requires a = sign in the password. Possibly also effects the username, as some pools require this too, but haven't tried that. Very tedious process, please look into this. At least now it doesn't hard lock the miner, but it simply doesn't pass the pool now.

Didn't know this existed, perhaps make it more visibly accessible, such as under the right click menu or main toolbar?

Also there is one really annoying persistent issue. Sometimes the miners will reorganize their positioning in the list. I don't know what causes it sometimes, but very annoying when it happens. Lock miner list does not prevent this from happening.

AM also doesn't remember column width between restarts

Lock miner list does not prevent repositioning of miners in the list in general.

Appreciate the work!

Couple Feat Reqs.

Please add the ability to query and display ASIC board models, PSUs, and miner model. Most modern firmware can display this. Such as on the summary.

Also add the ability to discern between temp diffs. Lets say if it sees two temps on two different boards, highlight two temps that are X degrees different then each other. This can either denote when a board is down or when it needs to be cleaned.

Feature Request: Highlight a row with a specific color. Such as a highlighter. Very useful for when you have a lot of units and you're trying to keep track of which ones are problematic, in maintenance, need to work on, or organize.

Might sound kinda dumb, but think it would help a lot.

Appreciate the continued updates.

Feature Req: Top bar for the group moves when you scroll down, so it's always at the top till you get to the next group, which bumps it out.

Feature Req: When you select multiple miners, the top bar is replaced with only the selected miners. Showing relevant data for only those, such as averages

Kind of a big deal here Patrike, it appears as though rules apply through disabled connections.

So if you disable connections rules don't respect the disabled state of miners. Also check DMs.

Got it.

Check your DMs.


Edit: Please FOR THE LOVE OF GOD add the ability to lock the columns, rows, spacing, sorting, and miner order. Locking the UI may seem like a small change, but oh god if you accidentally resort one of the columns it can literally destroy all of your workflow. Thanks.

Got it, not quite intuitive.

So to use it you go Managed Hosts > Local (which is your local client) > Auto discovery.

I was under the impression it was a standalone feature that just works on the client, but management is on a intranet level and you're looking at 'sites' under Managed Hosts, Local being the local client site.

I assume remote proxies would also show up in this list if I had any, which are other sites?

I guess I don't fully understand the context.

The Automatic Miner Discovery should just add new miners, right? But in order to use it, you have to click Add > then fill out criteria as if you were manually adding a new miner. You can't leave the remote agent IP address/hostname blank for instance. But that should be added, when automatically discovered? And would be different for each miner being added.

Hey Patrike, I appreciate you adding the ability to auto-add miners and perform a perpetual scan... Some things I've noticed, it appears as though it doesn't work properly? It's arranged under adding a new miner and you have to add criteria for miners you add? It feels as though this should be it's only function unrelated to manually adding a miner and maybe add custom criteria to incrementally change miners added, then permanently adjust them if they're ever changed via MAC address (so AM always knows which is which).

Also, while this has some technology overlap, when using MAC address identification please have the software set the IP address and remembered settings of the miner when it shows up again. Such as if the software or firmware changes. I understand you could use static addresses on your router, or with a active directory, however this combines the functionality all in one place and ties it specifically to AM. Makes overall management easier.

Using the autoscan and add function, you could also immediately assign a new miner to a IP range specifically for miners and have them statically present. Routers are not smart enough to understand what devices are being added to the network.

Couple things that still haven't been added to AM that it could really benefit from.

A way for AM to access a basic external API and retrieve/send data to use in variables. Very important for triggers or automating work. AM has it's own API, but can't communicate with external ones.

A 'WAIT' event for Triggers in rules. This allows you to set an amount of time in which rules need to be true (or false) before it triggers. Right now there is no way to do this and it just triggers.

Wanted to make sure I had this figured out. It looks like AM doesn't always function correctly and things are delayed or possibly aren't executed on systems that are heavily overloaded. The computer it was on was around 80-90% CPU utilization. Setting AM to high priority fixed this and haven't had a issue since doing this. I believe this issue can be duplicated by overloading the work system.

Could a option be added to run AM as high priority on startup (or whatever the person wants)?

Hey Patrike, running into some weird issues with Rules > Action > Wait. It seems like it doesn't trigger the first action and WAIT based on where it is in the list, it instead WAITs prior to doing any actions, at least with changing variables.

Please consider adding a 'wait' period to rules, triggers. So it'll wait before checking the next trigger if you have multiples. Right now for instance, if I have a variable setup and checking for hashrate for 60s, as soon as the variable is true (or whatever you have it setup to be), the other trigger is true and it'll immediately trigger instead of waiting to check hashrate despite wanting the hashrate check to happen after the variable changes.

It's available under actions, but not triggers.

Thanks.


I know you're trolling at this point so this is one of the last interactions we're going to have since you're wasting my time. 10pm comes after 9pm on a 12hour clock. Latest dates are at the top of the list. Patrike already fixed it, so appreciate that.

Sometimes posts get eaten here, thought you might have moderated rights to the thread.

TLDR:


1. Not true, there seems to be a problem with localization and while it's not the 0 position it's probably a issue with 12/24 hour format as it also has problems with the date, where it seems like it's not organized properly. Some parts of the world are DD/MM/YY, where the US is MM/DD/YY.

https://i.imgur.com/aMdiLL1.jpg

2. Static IPs are a lot of tedious work, especially as your network grows. The solution would be to run a active directory, which entails a lot more work. This is something that should be built into AM as AM is made to make mining easier, not harder, and I don't believe it's beyond scope.

3. Bandaid fix and not something you want in a mission critical environment. I have something similar already setup, but you have to manage each installation and make sure they're identical at all times or bad things happen.

Don't play dev when you're not the dev. Ideally we should be working to improve a piece of software, not come up with arbitrary rules as to why things can't be fixed or improved upon.


1. I didn't think of setting up longer leases (already done), although another bandaid I'll see if it helps. Ideally you don't want really long leases or you end up with devices that are no longer on your network eating a IP. Weirdly enough, the only time I've seen asics pull another IP is when the rule reboots them when they're locked up. They almost never pull new IPs or change.

2. Proxy looks to be a solution, although I feel as though this functionality should already be built into AM in some way. I intended on using a action to scan for new devices after the lockup rule goes off and readding them.

I already have MAC tracking turned on. For some reason after I readd them to AM they don't take over the original entry and are added as a new entry in the list.

Did my post get moderated or did it not post?

Appreciate looking into this, would this allow for a backup server then? Right now the client is the server and if the server goes offline, the problem still remains.


Couple other bugs I'm running into.

1. Tools > activity log > all

Appears as though the list isn't sorted or doesn't properly support 12 hour sorting. Also dates don't seem to be sorted properly due to them not properly parsing times that would have a 0 in front.

2. I'm running into a issue where IPs are changing on ASICs and the AM client doesn't update the current listing to reflect this. They can easily be found by scanning the network, but obviously not good. This is generally being caused by my router resetting then occasionally the miners resetting.

Could you add a Rule Action that runs a network scan and readds the miners replacing their original entry? They are uniquely identifiable at least by their MAC address.

Hey Patrike, considering how integral AM is to a lot of us at this point, could you look into adding failover. Like another installation of AM that will take over if the primary one goes down on your network.

Thanks

It seems as though there is a bug with the Time Rule Trigger. If you select a month range that is lets say late in the year till the next year, it doesn't properly read it. Like 10-04, so from Oct to April. It works fine in the opposite direction. So 1-8 or 4-12, but not 5-1 or whatever range you want that moves across the year till next year.