Backing up external HDD HTTP/2.0 connection failed

[Florius]

Hi,

I am unable to backup from 1 external HDD to another external HDD:

append chunks list len (64)
HTTP/2.0 connection failed
"Filepath"
catalog upload error - channel closed

I am however able to backup the root disk to the external HDD, and can also backup about 10 LXC containers.
Please advice, thank you!

 

[dietmar]

Maybe just a network failure. Have you tried multiple times?

 

[Florius]

Maybe just a network failure. Have you tried multiple times?

It's a LXC container, all happening on a local network (within the same server, even(. I tried 10 times, keeps giving the same error on different files.

 

[dietmar]

Anything special about that container? Please try to find out whats different to the other containers.

 

[Florius]

Anything special about that container? Please try to find out whats different to the other containers.

My apologies; the external HDD is connected to the server, on which runs a LXC container with another external HDD with passthrough.
So it happens from a physical machine to a container.
I can copy the internal SSD just fine though. I guess the problem is the external HDD, but why?

 

[dietmar]

My apologies; the external HDD is connected to the server, on which runs a LXC container with another external HDD with passthrough.
So it happens from a physical machine to a container.
I can copy the internal SSD just fine though. I guess the problem is the external HDD, but why?

Maybe you can share your LXC configuration? It quite hard to debug such things without any information...

Or even better: Provide us with the steps to create such container, so that we can reproduce the bug here.

 

[Florius]

Maybe you can share your LXC configuration? It quite hard to debug such things without any information...

Or even better: Provide us with the steps to create such container, so that we can reproduce the bug here.

Understandable, I just wasn't sure what information was required.
I create them with Terraform.

arch: amd64
cmode: tty
console: 1
cpulimit: 0
cpuunits: 1024
hostname: backup
memory: 512
mp0: /backup,mp=/backup
net0: name=eth0,bridge=vmbr0,hwaddr=0E:45:34:53:FB:82,ip=dhcp,type=veth
onboot: 0
ostype: debian
protection: 0
rootfs: local-lvm:vm-109-disk-0,size=10G
swap: 512
tty: 2

No clue if it's useful, but whatever:

terraform {
  required_providers {
    proxmox = {
      source = "Telmate/proxmox"
      version = "2.7.0"
    }
  }
}


provider "proxmox" {
    pm_api_url = "https://server-01/api2/json"
}


locals {
  containers = [
    { hostname = "unifi", template = "local:vztmpl/debian-9.0-standard_9.7-1_amd64.tar.gz", memory = "2048", grains = { roles = ["unifi", "haproxy", "mongodb"] } },
    { hostname = "icinga", template = "local:vztmpl/debian-10-standard_10.5-1_amd64.tar.gz", grains = { roles = ["icinga-master", "mysql", "redis", "apache"]} },
    { hostname = "pihole", template = "local:vztmpl/debian-10-standard_10.5-1_amd64.tar.gz", ip = "192.168.1.253/24", gw = "192.168.1.1", grains = { roles = ["pihole", "lighttpd"]} },
    { hostname = "prometheus", template = "local:vztmpl/debian-10-standard_10.5-1_amd64.tar.gz", grains = { roles = ["prometheus", "haproxy"]} },
    { hostname = "grafana", template = "local:vztmpl/debian-10-standard_10.5-1_amd64.tar.gz", size = "30G", grains = { roles = ["influxdb", "grafana", "haproxy"]} },
    { hostname = "librenms", template = "local:vztmpl/debian-10-standard_10.5-1_amd64.tar.gz" , grains = { roles = ["librenms", "mysql", "apache"]} },
    { hostname = "smokeping", template = "local:vztmpl/debian-10-standard_10.5-1_amd64.tar.gz" , grains = { roles = ["smokeping", "apache"]} },
    { hostname = "mqttbroker", template = "local:vztmpl/debian-10-standard_10.5-1_amd64.tar.gz", grains = { roles = ["mqttbroker", "haproxy"] } },
    { hostname = "homeassistant", template = "local:vztmpl/ubuntu-20.04-standard_20.04-1_amd64.tar.gz", grains = { roles = ["homeassistant", "haproxy"] } },
    { hostname = "backup", template = "local:vztmpl/debian-10-standard_10.5-1_amd64.tar.gz", grains = { roles = ["backup"] } },
  ]
}


resource "proxmox_lxc" "create_container" {
  count = length(local.containers)


  vmid = 100+count.index
  hostname = lookup(local.containers[count.index], "hostname")
  ostemplate = lookup(local.containers[count.index], "template")
  target_node = "server-01"
  ostype = "debian"
  start = true
  memory = lookup(local.containers[count.index], "memory", "512")
  rootfs {
    storage = lookup(local.containers[count.index], "storage", "local-lvm")
    size = lookup(local.containers[count.index], "size", "10G")
  }


  network {
    name   = "eth0"
    bridge = "vmbr0"
    ip     = lookup(local.containers[count.index], "ip", "dhcp")
    gw     = lookup(local.containers[count.index], "gw", null)
  }


  provisioner "local-exec" {
    when    = destroy
    command = <<EOT
      sudo salt-key --force-color --yes -d ${self.hostname}
      EOT
  }


  provisioner "local-exec" {
    command = <<EOT
      sleep 5s
      sudo pct exec ${count.index+100} -- /bin/bash -c "apt update && apt install -y curl sudo"
      sudo pct exec ${count.index+100} -- /bin/bash -c "curl -L https://bootstrap.saltproject.io | sudo sh -s -- -X -x python3 stable 3003"
      sudo pct exec ${count.index+100} -- /bin/bash -c "systemctl stop salt-minion"
      sudo pct exec ${count.index+100} -- /bin/bash -c "echo 'master: server-01 > /etc/salt/minion.d/99-master-address.conf"
      sudo lxc-attach ${count.index+100} -- /bin/bash -c "echo '${yamlencode(lookup(local.containers[count.index], "grains"))}' > /etc/salt/grains"
      sudo pct exec ${count.index+100} -- /bin/bash -c "systemctl restart salt-minion"
      sleep 5s
      sudo salt-key --force-color --yes -a  ${lookup(local.containers[count.index], "hostname")}
    EOT
  }
}

Please let me know what else I can provide to help!

EDIT: Added a screenshot, it also just doesn't complete in the UI.
There are no more running task, and the status is "stopped: unknown".
The last few log lines from the UI:

2021-05-24T09:09:08+00:00: POST /dynamic_chunk
2021-05-24T09:09:08+00:00: upload_chunk done: 3245323 bytes, c9c21db43c0d61afb3d9fdaddeb9181cb858c1ed1cd5ad6e4f96cad4140d5724
2021-05-24T09:09:08+00:00: upload_chunk done: 5053531 bytes, cb759c64d8c5a60a363c256c04589f81120d916b2993947faf535ef7ca8779d3
2021-05-24T09:09:09+00:00: POST /dynamic_chunk
2021-05-24T09:09:09+00:00: upload_chunk done: 2433474 bytes, 60e30a96c5ebc4d3ed9ace2fbadd24ae8e1f6b0c9218535fdede539ce8c6441e
2021-05-24T09:09:09+00:00: upload_chunk done: 2135478 bytes, d299a3e97c1728127080a359afcbefdee731ded7f7518dfde003903e68844ccf
2021-05-24T09:09:09+00:00: POST /dynamic_chunk
2021-05-24T09:09:09+00:00: POST /dynamic_chunk

EDIT2: It are non-native Linux files though, like .exe, .rar, .bin fils and such. Could this have anything to do with it?

Thanks!

 

[dietmar]

What kind of file systen is /backup. Does backup work if you (temporaily) remove that mount point?

 

[dietmar]

Also, please make sure you run the latest version...

 

[Florius]

Also, please make sure you run the latest version...

Thanks Dietmar for your advice.

I run version 1.1.7-1 as client, and 1.1-5 as server.
The file system is ext4.

I'm afraid I don't understand your comment regarding the unmount.
I unmounted on the server side, and tried again, resulting in the same error.

 

[Florius]

Does anyone have any other suggestions I could try?

Thank you!