Cannot complete first run when using small tmpfs for /tmp: "No space left on device" error

Steps to reproduce

1. Create a Fedora 23 machine with 1GB RAM. This, by default, creates /tmp using the in-memory tmpfs set to 500 MB.
2. Run dotnet (anything) to trigger first run nuget cache priming

Expected behavior

Prime nuget cache succeeds

Actual behavior

This fills /tmp and crashes. There is no obvious way for user to opt-out of first run.

A command is running to initially populate your local package cache, to improve restore speed and enable offline access. This command will take up to a minute to complete and will only happen once.
Decompressing 100% 2628 ms
Expanding 74%
Unhandled Exception: System.AggregateException: One or more errors occurred. (One or more errors occurred. (No space left on device)) (One or more errors occurred. (No space left on device)) (No space left on device) (No space left on device) ---> System.IO.IOException: No space left on device
   at System.IO.UnixFileStream.CheckFileCall(Int64 result, Boolean ignoreNotSupported)
   at System.IO.UnixFileStream.WriteNative(Byte[] array, Int32 offset, Int32 count)
   at System.IO.UnixFileStream.FlushWriteBuffer()
   at System.IO.UnixFileStream.Dispose(Boolean disposing)
   at System.IO.FileStream.Dispose(Boolean disposing)
   at System.IO.Stream.Close()
   at System.IO.Compression.ZipArchive.CloseStreams()
   at System.IO.Compression.ZipArchive.Dispose(Boolean disposing)
   at System.IO.Compression.ZipArchive.Dispose()
   at Microsoft.DotNet.Archive.IndexedArchive.ZipOperation.Execute()
   at Microsoft.DotNet.Archive.IndexedArchive.ExtractOperation.DoOperation()
   at Microsoft.DotNet.Archive.IndexedArchive.<>c__DisplayClass19_1.<Extract>b__0(ExtractOperation extractOperation)
   at System.Linq.Parallel.ForAllOperator`1.ForAllEnumerator`1.MoveNext(TInput& currentElement, Int32& currentKey)
   at System.Linq.Parallel.ForAllSpoolingTask`2.SpoolingWork()
   at System.Linq.Parallel.SpoolingTaskBase.Work()
   at System.Linq.Parallel.QueryTask.BaseWork(Object unused)
   at System.Linq.Parallel.QueryTask.RunTaskSynchronously(Object o)
   at System.Threading.Tasks.Task.Execute()
   --- End of inner exception stack trace ---
   at System.Linq.Parallel.QueryTaskGroupState.QueryEnd(Boolean userInitiatedDispose)
   at System.Linq.Parallel.MergeExecutor`1.Execute()
   at System.Linq.Parallel.MergeExecutor`1.Execute[TKey](PartitionedStream`2 partitions, Boolean ignoreOutput, ParallelMergeOptions options, TaskScheduler taskScheduler, Boolean isOrdered, CancellationState cancellationState, Int32 queryId)
   at System.Linq.Parallel.PartitionedStreamMerger`1.Receive[TKey](PartitionedStream`2 partitionedStream)
   at System.Linq.Parallel.ForAllOperator`1.WrapPartitionedStream[TKey](PartitionedStream`2 inputStream, IPartitionedStreamRecipient`1 recipient, Boolean preferStriping, QuerySettings settings)
   at System.Linq.Parallel.UnaryQueryOperator`2.UnaryQueryOperatorResults.ChildResultsRecipient.Receive[TKey](PartitionedStream`2 inputStream)
   at System.Linq.Parallel.ListQueryResults`1.GivePartitionedStream(IPartitionedStreamRecipient`1 recipient)
   at System.Linq.Parallel.UnaryQueryOperator`2.UnaryQueryOperatorResults.GivePartitionedStream(IPartitionedStreamRecipient`1 recipient)
   at System.Linq.Parallel.QueryOperator`1.GetOpenedEnumerator(Nullable`1 mergeOptions, Boolean suppressOrder, Boolean forEffect, QuerySettings querySettings)
   at System.Linq.Parallel.ForAllOperator`1.RunSynchronously()
   at System.Linq.ParallelEnumerable.ForAll[TSource](ParallelQuery`1 source, Action`1 action)
   at Microsoft.DotNet.Archive.IndexedArchive.Extract(String compressedArchivePath, String outputDirectory, IProgress`1 progress)
   at Microsoft.DotNet.Configurer.NuGetPackagesArchiver.ExtractArchive()
   at Microsoft.DotNet.Configurer.NuGetCachePrimer.PrimeCache()
   at Microsoft.DotNet.Configurer.DotnetFirstTimeUseConfigurer.Configure()
   at Microsoft.DotNet.Cli.Program.ConfigureDotNetForFirstTimeUse(INuGetCacheSentinel nugetCacheSentinel)
   at Microsoft.DotNet.Cli.Program.ProcessArgs(String[] args, ITelemetry telemetryClient)
   at Microsoft.DotNet.Cli.Program.Main(String[] args)

After run:

$ df -h 
Filesystem                                     Size  Used Avail Use% Mounted on
tmpfs                                          495M  495M  4.0K 100% /tmp

Environment data

dotnet --info output:

.NET Command Line Tools (1.0.0-preview2-003118)

Product Information:
 Version:            1.0.0-preview2-003118
 Commit SHA-1 hash:  7ada53d523

Runtime Environment:
 OS Name:     fedora
 OS Version:  23
 OS Platform: Linux
 RID:         fedora.23-x64

Changes to consider

• Put the env variable in first run message so users more easily find out how to skip this
• Allow specifying which directory to use during extraction
• Cleanup tmp dir if extraction fails
• Run extraction into nuget cache directly instead of into tmp dir first.

Workaround

Set environment variable DOTNET_SKIP_FIRST_TIME_EXPERIENCE=true permanently (e.g. add to .bashrc or something).

[omajid]

Could you elaborate on how I would “Set environment variable DOTNET_SKIP_FIRST_TIME_EXPERIENCE=true permanently (e.g. add to .bashrc or something).”

This depends on what shell you are using. If you don’t know, it is probably bash. For that, add the following line to your ~/.bashrc file. This is a file named .bashrc in your home directory. If the file doesn’t exist, create it.

export DOTNET_SKIP_FIRST_TIME_EXPERIENCE=true

Then log out and login again. You can verify that this worked by doing:

env | grep DOTNET

You should see:

DOTNET_SKIP_FIRST_TIME_EXPERIENCE=true

When you run dotnet now, you shouldn’t see the first run message or the error.

[livarcocc]

We have removed the lzma cache expansion logic from the 3.0 CLI already. Closing this issue.