JonathanM

That wasn't quite the point. Do those extra drives show up in /mnt at the unraid terminal after the docker is running? Whether or not they show up as drives to Zabbix isn't material to the question asked. The issue is that if for instance you don't have a /mnt/disk10, when you start the docker with a /mnt/disk10 host path defined it can create /mnt/disk10 on the host, but there isn't an actual disk there, so it ends up being a path in RAM, which looks and acts similar to a real drive to someone perusing the file system on the host unraid.

Yes, I disabled that before I updated the container. I have no clue what happens when a docker restart is issued right in the middle of an update. 😀

Edit the vm and change from form view to xml view, and remove the section that references the device from the xml

@Conmyster,Edit the first post of this thread and change the title to something that better reflects the content. Something like Zabbix Agent (solved, docker instructions here) or something like that. Possibly also @trurl may want to move the thread to a different subforum, since feature requests is definitely no longer the proper place for it.

New :test downloaded. Keep in mind I don't generally interact with this, everything is automated. So it could be days before we get a result, weeks before it can be called fixed. I'll keep an eye on it regularly, but there is nothing in the queue at the moment.

That card uses a marvell controller, NOT recommended. I think amazon B006SF68OS would be a better pick.

For those of us who have never heard of Zabbix, perhaps it would be in your best interest to describe things a bit more. If you put forward a compelling case as to why we should want it, you may get more interest. My first thought is "Why should I care about Zabbix?" If you can answer that in a way that makes someone want to start using it, you will get a better response. Or, perhaps I'm just a luddite and everyone who is cool already knows and uses Zabbix. In that case, never mind.

If you would have read the thread, they discuss deleting the file after the array has been started.

Tools, diagnostics. Attach the zip file you download from there intact to your next post.

Have a read here. This has been discussed many times in the past. https://forums.unraid.net/topic/61973-encryption-and-auto-start/

This issue is effecting a wide swath of people. Personally I have 27 containers on one of my servers, all 10 of the LSIO show update ready, none of the other containers do. They were all up to date yesterday. No connection issues at my end of the pipe, one of my other servers behind the same router just updated a binhex container just fine. A different server at another location with 10 containers, all 3 of the LSIO show update ready. If it's unraid's fault, then unraid is preferentially picking on LSIO.

This response 2 months ago in this thread sums it up nicely. https://forums.unraid.net/topic/51808-support-linuxserverio-letsencrypt-nginx/?do=findComment&comment=748653

Why does it only effect LSIO containers? Did unraid blacklist you?🤣

USB, while it may work, is extremely problematic for array drives. It's just not a robust enough protocol for the demanding environment of multi disk RAID, or unraid. Did you originally have all your drives directly connected to SATA ports? Describe your hardware in much more detail, and maybe we can find a solution that will work.

LOL, on point!

Suffice it to say, it's a sore subject. Been discussed to death in the forums. Caused many issues, created many arguments. First rule of unifi thread, don't mention unifi updates. We love unifi hardware, hate unifi half done updates that break great hardware.

Docker compose is available in nerdpack. https://forums.unraid.net/topic/35866-unraid-6-nerdpack-cli-tools-iftop-iotop-screen-kbd-etc/?do=findComment&comment=653719

Yes. I use the LTS tag. But look at it from LSIO's viewpoint. If they push the bleeding edge and something goes wrong, which to be fair, stable for unifi IS bleeding edge, they get to hear how everybody's install is borked. If you use their container, it's up to them how soon they update. You are free to make the decision as a unifi user to run the absolute latest code the second it comes out if you wish, just not in LSIO's container.

Which may or may not mean it's a good idea to push that version to a production environment. "Stable" unifi software has caused major headaches in the past, I'd much rather wait until it's been running on someone else's system for a while before I trust my multiple sites to it. If wifi goes down, it's a big deal. I'd rather not deal with angry users.

I doubt that very much. Air follows the path of least resistance. If the intake fans are being properly ducted so none of the air can bypass direct contact with the drives, the number of fans doesn't much matter. It takes a lot of force to push air through the small passages beside the drives, so if there is an easier path bypassing the drives, that's where the air will go. It's your job to constrain that air and make it go where you need it to move the heat. In a server with a load of drives, the first priority is the drive wall, then the CPU, hard drive controller, memory, motherboard, and GPU. If you don't duct the flow where you need it, you will end up with a lot of flow that bypasses all the critical components, and pockets of heat that don't escape.

If the hole doesn't have a fan forcing air out, it needs to be blocked off. Otherwise your exhaust fans will just be getting make up air from the open holes instead of pulling air over the drives. The only place that should allow incoming air is through the drives. Clear packing tape is good for sealing up holes.

Possible bug report. Not a show stopper, just unexpected behaviour. When I reduced the number of workers and applied the setting, the extraneous workers finished their job, but the file was not replaced in the library, nor was the temporary location cleared. I restarted the container, and the temporary work was deleted and the original files were untouched. The files in question were subsequently converted. Not a huge issue, just means that if you reduce the number of workers you will lose the time they spent on their current job.

Still actively using it, ALL members of the household must be invested in keeping it current, or it doesn't work, like you said. It's just me and the wife, and she sees the value of keeping it up to date because it makes shopping much quicker. Most interactions are on our phones, I only used a full size browser during initial inventory load. Making dinner? grab phone, click on grocy, scroll down stock overview screen and tap on everything we opened. I don't track opened and partial things as a general rule, if the package is opened, it's out of inventory. I also don't track stuff that doesn't stick around more than a week, like bread. The inventory list is purely reserve stock, nothing currently being consumed. Now, I only track items that I know through past experience are an issue when shopping, I don't track everything. For example, we shop at bulk stores typically 5 or 6 times each year, and it's really easy to forget you purchased a 3 pack of ketchup last time, and end up with 7 unopened bottles. As I type this, I have 87 line items with 665 total units tracked. I can easily think of at least 20 more consumable items that are not tracked, simply because it's not worth the effort. If they run out and I don't remember to purchase replacements, I didn't need them anyway. Luxury items like ice cream, candy, canned drinks, etc. This app is great for people with the correct amount of OCD tendency. (no relation to true medically diagnosed OCD, I know the difference and am eternally grateful I'm not) You need to care enough to be motivated to keep the counts close, but not so OCD that every item on the shelf is entered and tracked.

Cue George Takei "Ohhh Myyyy!"