Open wmertens opened 4 years ago
So, How to solve this problem, Does anyone idea here?
Can you share more information about the issue? Does the error occur when you launch a 'nixops deploy' operation? Can you check if you have enough space in the deployment server itself? Can you try to run a nix-collect-garbage to free up some space from the nix store?
Can you share more information about the issue? Does the error occur when you launch a 'nixops deploy' operation?
machine setting:
{
photo =
{ config, pkgs, lib, ... }:
{
deployment.targetEnv = "virtualbox";
deployment.virtualbox.memorySize = 6028; # megabytes
deployment.virtualbox.vcpu = 4; # number of cpus
virtualisation.virtualbox.guest.enable = lib.mkForce false;
boot.tmpOnTmpfs = true;
};
}
check machine info:
nixops check -d photo
Machines state:
+-------+--------+-----+-----------+----------+----------------+------------------------------------------+-------+
| Name | Exists | Up | Reachable | Disks OK | Load avg. | Units | Notes |
+-------+--------+-----+-----------+----------+----------------+------------------------------------------+-------+
| photo | Yes | Yes | Yes | N/A | 0.39 0.28 0.16 | ● home.mount [failed] | |
| | | | | | | sys-fs-fuse-connections.mount [failed] | |
| | | | | | | sys-kernel-config.mount [failed] | |
+-------+--------+-----+-----------+----------+----------------+------------------------------------------+-------+
Non machines resources state:
+------+--------+
| Name | Exists |
+------+--------+
+------+--------+
nixops ssh -d photo photo (build package test failed!)
nix-shell -p git
git clone https://github.com/GTrunSec/photoprism-flake
cd photoprism-flake/
nix-build
error output:
go: downloading github.com/dsoprea/go-photoshop-info-format v0.0.0-20200610045659-121dd752914dgo: downloading github.com/go-xmlfmt/xmlfmt v0.0.0-20191208150333-d5b6f63a941b-> unzip /build/go/pkg/mod/cache/download/github.com/tensorflow/tensorflow/@v/v1.15.2.zip: write /build/go/pkg/mod/github.com/tensorflow/tensorflow@v1.15.2.tmp-911731347/tensorflow/python/kernel_tests/sparse_conditional_accumulator_test.py: no space left on device
github.com/photoprism/photoprism/internal/classify imports
github.com/tensorflow/tensorflow/tensorflow/go: unzip /build/go/pkg/mod/cache/download/github.com/tensorflow/tensorflow/@v/v1.15.2.zip: write /build/go/pkg/mod/github.com/tensorflow/tensorflow@v1.15.2.tmp-911731347/tensorflow/python/kernel_tests/sparse_conditional_accumulator_test.py: no space left on device
github.com/photoprism/photoprism/internal/nsfw imports
github.com/tensorflow/tensorflow/tensorflow/go/op: unzip /build/go/pkg/mod/cache/download/github.com/tensorflow/tensorflow/@v/v1.15.2.zip: write /build/go/pkg/mod/github.com/tensorflow/tensorflow@v1.15.2.tmp-911731347/tensorflow/python/kernel_tests/sparse_conditional_accumulator_test.py: no space left on device
df check
df -ih
Filesystem Inodes IUsed IFree IUse% Mounted on
/dev/disk/by-label/nixos 3.2M 210K 3.0M 7% /
tmpfs 732K 18 732K 1% /sys/fs/cgroup
devtmpfs 730K 379 730K 1% /dev
tmpfs 732K 1 732K 1% /dev/shm
tmpfs 732K 898 731K 1% /run
tmpfs 732K 39 732K 1% /run/wrappers
tmpfs 147K 15 147K 1% /run/user/0
tmpfs 732K 10 732K 1% /tmp
Thanks for your helping.
solved by resizing the tmp
Cool!
cc @wmertens did you face the exact same issue?
yes but I solved it by setting TMPDIR=/tmp
. I don't have tmpOnTmpfs
.
Perhaps there should be a check the available disk space on $TMPDIR
, if it's less than e.g. 500MB print a warning?
Running nixops as regular user id 1000 with these specs:
causes the opaque error
tar: no space left on device
The reason is that nixops uses
TMPDIR
which points to /run and is limited to 100M.What would be a good way to tell the user about this problem?