Unify CLI options (verbosity, version) #685

mkesper · 2025-01-31T14:53:49Z

Feel free to adjust block size. But 1k blocks are ridiculously low for downloading GiB of LLM data.

Related: #684

Summary by Sourcery

New Features:

Added a --quiet flag to suppress the progress bar during downloads.

sourcery-ai · 2025-01-31T14:53:55Z

Reviewer's Guide by Sourcery

This pull request introduces a new --quiet flag to suppress progress bars during downloads, and increases the download block size to 100KB to improve download speeds and reduce CPU usage.

Sequence diagram for the modified download process

sequenceDiagram
    participant Client
    participant DownloadManager
    participant HTTPClient

    Client->>DownloadManager: download_file(url, show_progress)
    DownloadManager->>HTTPClient: init(url, show_progress)
    activate HTTPClient
    loop Until EOF
        HTTPClient->>HTTPClient: read(100KB blocks)
        Note right of HTTPClient: Increased from 1KB to 100KB blocks
        alt show_progress is true
            HTTPClient->>Client: Update progress bar
        end
    end
    HTTPClient-->>DownloadManager: Download complete
    deactivate HTTPClient
    DownloadManager-->>Client: File downloaded

Class diagram showing modified download components

classDiagram
    class HTTPClient {
        -response
        -now_downloaded: int
        -start_time: float
        +init(url, headers, output_file, show_progress)
        +perform_download(file, progress)
    }

    class DownloadManager {
        +download_file(url, dest_path, headers, show_progress)
    }

    class PullCommand {
        +pull(args)
    }

    PullCommand ..> DownloadManager : uses
    DownloadManager ..> HTTPClient : uses

    note for HTTPClient "Block size increased to 100KB"

File-Level Changes

Change	Details	Files
Added a `--quiet` flag to suppress progress bars during downloads.	Added a `--quiet` argument to the `pull_parser` in `ramalama/cli.py`. Modified `ramalama/url.py` to pass the `quiet` argument to the `download_file` function. Modified `ramalama/oci.py` to pass the `quiet` argument to the container engine.	`ramalama/cli.py` `ramalama/url.py` `ramalama/oci.py`
Increased the download block size to 100KB.	Changed the read size in `perform_download` from 1024 bytes to 100 * 1024 bytes.	`ramalama/http_client.py`
Modified `download_file` to accept a `show_progress` argument.	Modified `download_file` in `ramalama/common.py` to accept a `show_progress` argument. Modified `init_pull` and `pull_blob` in `ramalama/ollama.py` to pass the `show_progress` argument to `download_file`.	`ramalama/common.py` `ramalama/ollama.py`

Tips and commands

Interacting with Sourcery

Trigger a new review: Comment @sourcery-ai review on the pull request.
Continue discussions: Reply directly to Sourcery's review comments.
Generate a GitHub issue from a review comment: Ask Sourcery to create an
issue from a review comment by replying to it. You can also reply to a
review comment with @sourcery-ai issue to create an issue from it.
Generate a pull request title: Write @sourcery-ai anywhere in the pull
request title to generate a title at any time. You can also comment
@sourcery-ai title on the pull request to (re-)generate the title at any time.
Generate a pull request summary: Write @sourcery-ai summary anywhere in
the pull request body to generate a PR summary at any time exactly where you
want it. You can also comment @sourcery-ai summary on the pull request to
(re-)generate the summary at any time.
Generate reviewer's guide: Comment @sourcery-ai guide on the pull
request to (re-)generate the reviewer's guide at any time.
Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
pull request to resolve all Sourcery comments. Useful if you've already
addressed all the comments and don't want to see them anymore.
Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
request to dismiss all existing Sourcery reviews. Especially useful if you
want to start fresh with a new review - don't forget to comment
@sourcery-ai review to trigger a new review!
Generate a plan of action for an issue: Comment @sourcery-ai plan on
an issue to generate a plan of action for it.

Customizing Your Experience

Access your dashboard to:

Enable or disable review features such as the Sourcery-generated pull request
summary, the reviewer's guide, and others.
Change the review language.
Add, remove or edit custom review instructions.
Adjust other review settings.

Getting Help

Contact our support team for questions or feedback.
Visit our documentation for detailed guides and information.
Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

sourcery-ai

Hey @mkesper - I've reviewed your changes - here's some feedback:

Overall Comments:

The --quiet flag should use action="store_true" instead of action="store" with default=False to be consistent with other boolean flags in the codebase

Here's what I looked at during the review

🟡 General issues: 1 issue found
🟢 Security: all looks good
🟢 Testing: all looks good
🟢 Complexity: all looks good
🟢 Documentation: all looks good

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

sourcery-ai · 2025-01-31T14:54:47Z

ramalama/ollama.py

@@ -45,9 +45,9 @@ def pull_blob(repos, layer_digest, accept, registry_head, models, model_name, mo
    run_cmd(["ln", "-sf", relative_target_path, model_path])


-def init_pull(repos, accept, registry_head, model_name, model_tag, models, model_path, model):
+def init_pull(repos, accept, registry_head, model_name, model_tag, models, model_path, model, show_progress):


issue (bug_risk): Function signature mismatch: pull_config_blob is called with show_progress parameter but doesn't accept it

ericcurtin · 2025-02-01T10:17:13Z

Change looks fine to me @mkesper you just need to sign your commits with something like:

git commit -a -s --amend

to satisfy the DCO bot.

ericcurtin · 2025-02-01T10:50:44Z

Lots of tests failed too, we need to get this build green somehow, at least some of the failures are flakes, not sure how many.

This will all be rebuilt anyway when you the commits are signed.

jhjaggars · 2025-02-01T16:28:17Z

ramalama/common.py

@@ -177,7 +177,7 @@ def download_file(url, dest_path, headers=None, show_progress=True):
        show_progress = False

    try:
-        http_client.init(url=url, headers=headers, output_file=dest_path, progress=show_progress)
+        http_client.init(url=url, headers=headers, output_file=dest_path, show_progress=show_progress)


This rename needs to be made to HttpClient.init as well.

ericcurtin · 2025-02-01T16:57:23Z

Anybody know what "podman pull" or "podman artifact pull" block size is? @baude @rhatdan

We could just standardise on that, it's likely a sane value.

I probably care about reliability over cpu usage. I know in practice at the TCP level of the stack a max packet size is 1500 bytes, but that's probably not so important at the http level.

ericcurtin · 2025-02-01T16:59:04Z

It could be the progress bar causing the CPU usage issue also, maybe we could change it so that it doesn't update every single block iteration.

ericcurtin · 2025-02-01T17:30:31Z

Does quiet alone lower the CPU usage significantly @mkesper ?

jhjaggars · 2025-02-03T01:00:04Z

@ericcurtin Best I can tell, the image copier used in podman updates once per second by default:
https://github.com/containers/podman/blob/main/vendor/github.com/containers/common/libimage/copier.go#L315

ericcurtin · 2025-02-03T01:55:44Z

If you guys could check if this makes a difference I'd appreciate it, not much CPU usage issues on my machine (and maybe the time calculation is as expensive that printing a progress bar update, I've no idea):

#717

swarajpande5 · 2025-02-04T15:27:58Z

ramalama/ollama.py

@@ -26,11 +26,11 @@ def pull_config_blob(repos, accept, registry_head, manifest_data):
    download_file(url, config_blob_path, headers=headers, show_progress=False)


ramalama/ramalama/ollama.py

Lines 18 to 26 in 94bd95e

def pull_config_blob(repos, accept, registry_head, manifest_data):

cfg_hash = manifest_data["config"]["digest"]

config_blob_path = os.path.join(repos, "blobs", cfg_hash)

os.makedirs(os.path.dirname(config_blob_path), exist_ok=True)

url = f"{registry_head}/blobs/{cfg_hash}"

headers = {"Accept": accept}

download_file(url, config_blob_path, headers=headers, show_progress=False)

Please add the show_progress parameter in this function and use the same in the download_file() call, removing the hard coded False.

I can do this to make it more consistent although this will change behaviour to before the patch.

I think False is always correct here, it's not a bulky file, the progress means very little, we only show progress for the gguf file

sounds good !

mkesper · 2025-02-05T21:17:23Z

Does quiet alone lower the CPU usage significantly @mkesper ?

Sorry, I had no time to check yet.

mkesper · 2025-02-06T17:09:38Z

I have updated the commits after testing them. With the higher block size I do not get any relevant CPU usage shown for terminal, even without --quiet. With the changes I get about 15% on one CPU thread with progress and about 10% with --quiet turned on. Before, it was like in the screenshot when pulling fro ollama: about 50% of a CPU thread for the Python process, about 30% for tmux and additional 25% for konsole (terminal application).

mkesper · 2025-02-06T17:11:59Z

To add: Interrupting the download is still perfectly possible with Ctrl-C.

rhatdan · 2025-02-07T10:23:41Z

Thanks @mkesper
LGTM

dougsland · 2025-02-07T12:01:19Z

ramalama/http_client.py

@@ -60,7 +60,7 @@ def perform_download(self, file, progress):
        self.now_downloaded = 0
        self.start_time = time.time()
        while True:
-            data = self.response.read(1024)
+            data = self.response.read(100 * 1024)


The patch looks good to me. How about we add a constant 100KB=100*1024 and use it here? It's faster reading for all people.

I'm not sure this is the best fix, I think this works because the read is so big it delays the time that the next iteration of the loop is called. In practice you cannot depend on TCP packet sizes being larger than 1400 bytes (because of middleboxes, legacy, etc.) and this encourages fragmenting the bytes, which isn't spectacular for reliability, performance (from the transport perspective), etc.

I think it's likely the whole printing the progress bar thing is contributing to spiking the CPU usage, so if we just avoid updating that I think it could be a better fix to try (just updated this PR):

#717

But I don't have CPU spike issues personally so cannot test the above PR.

But reading python http library, they seem to default to 8kb, that seems more reasonable than 100kb:

https://github.com/python/cpython/blob/main/Lib/http/client.py

We could try both PRs together

I'm happy to drop #717 if it doesn't make much difference to CPU usage, although it does seem excessive to update the progress bar on each iteration.

I'm not sure this is the best fix, I think this works because the read is so big it delays the time that the next iteration of the loop is called. In practice you cannot depend on TCP packet sizes being larger than 1400 bytes (because of middleboxes, legacy, etc.) and this encourages fragmenting the bytes, which isn't spectacular for reliability, performance (from the transport perspective), etc.

As far as I understand the Python http client library code, read() does not operate at TCP level but at HTTP stream level, right? That would render this concern moot.

True, but it's still a consideration as HTTP is on top of TCP

IMHO, Splitting in two patches make sense, one for the progress bar and another for buffer size to be discussed. The progress bar skipping should already help as @ericcurtin mentioned.

dougsland · 2025-02-07T12:02:04Z

LGTM I add minor comment.

dougsland · 2025-02-07T12:02:56Z

CI/CD is complaning about:

xref-helpmsgs-manpages: 'ramalama pull --help' lists '--quiet', which is not in docs/ramalama-pull.1.md
make: *** [Makefile:133: validate] Error 1

dougsland · 2025-02-07T12:04:16Z

The another issue in CI/CD don't see related to the patch:

# tags: distro-integration
# (from function `bail-now' in file test/system/helpers.podman.bash, line 122,
#  from function `die' in file test/system/helpers.podman.bash, line 848,
#  from function `run_ramalama' in file test/system/helpers.bash, line 187,
#  in test file test/system/055-convert.bats, line 12)
#   `run_ramalama 1 convert bogus foobar' failed
#
# [10:29:52.5[68](https://github.com/containers/ramalama/actions/runs/13183889748/job/36843129951?pr=685#step:5:69)712822] $ /home/runner/work/ramalama/ramalama/bin/ramalama convert
# [10:29:52.667041482] usage: ramalama convert [-h] [--type {car,raw}] [--network-mode NETWORK_MODE]
#                         SOURCE TARGET
# ramalama convert: error: the following arguments are required: SOURCE, TARGET
# [10:29:52.6[70](https://github.com/containers/ramalama/actions/runs/13183889748/job/36843129951?pr=685#step:5:71)168914] [ rc=2 (expected) ]
#
# [10:29:52.681213373] $ /home/runner/work/ramalama/ramalama/bin/ramalama convert tiny
# [10:29:52.777658054] usage: ramalama convert [-h] [--type {car,raw}] [--network-mode NETWORK_MODE]
#                         SOURCE TARGET
# ramalama convert: error: the following arguments are required: TARGET
# [10:29:52.7809198[75](https://github.com/containers/ramalama/actions/runs/13183889748/job/36843129951?pr=685#step:5:76)] [ rc=2 (expected) ]
#
# [10:29:52.792616049] $ /home/runner/work/ramalama/ramalama/bin/ramalama convert bogus foobar
# [10:29:52.892675855] usage: ramalama [-h] [--container] [--debug] [--dryrun] [--engine ENGINE]
#                 [--gpu] [--ngl NGL] [--keep-groups] [--image IMAGE]
#                 [--nocontainer] [--runtime {llama.cpp,vllm}] [--store STORE]
#                 [-v]
#                 {help,bench,benchmark,containers,ps,convert,info,list,ls,login,logout,perplexity,pull,push,rm,run,serve,stop,version}
#                 ...
# ramalama: requires a subcommand
# #/vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv
# #| FAIL: exit code is 0; expected 1
# #\^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
# # [teardown]
ok 34 [055] ramalama convert file to image in 3116ms
ok 35 [055] ramalama convert tiny to image in 16685ms
ok 36 [060] ramalama info in 16[83](https://github.com/containers/ramalama/actions/runs/13183889748/job/36843129951?pr=685#step:5:84)ms
make: *** [Makefile:149: bats-nocontainer] Error 1e

rhatdan · 2025-02-07T14:51:30Z

Was this a duplicate of the other PR?

ericcurtin · 2025-02-07T15:04:00Z

They are related, we can take this PR in too if the tests are fixed so it's passing CI. 100kb is an excessively high block size though, we'd have to reduce that

ericcurtin · 2025-02-07T15:05:33Z

High block sizes can break some network connections outright...

rhatdan · 2025-02-13T13:28:18Z

The add subcommand means most likely that --quiet option was not added to one of the parsers.

Related: containers#685 Signed-off-by: Michael Kesper <[email protected]>

rhatdan · 2025-02-13T21:22:24Z

Squash and sign your commits.

Related: containers#685 Signed-off-by: Michael Kesper <[email protected]>

mkesper · 2025-02-15T16:35:32Z

Seems I'm stuck now, can't see how to fix this (seems to be connected to autocomplete):

./hack/xref-helpmsgs-manpages
xref-helpmsgs-manpages: 'ramalama  --help' lists '--quiet,', which is not in docs/ramalama.1.md
xref-helpmsgs-manpages: 'ramalama ': '--quiet' in docs/ramalama.1.md, but not in --help```

Related: containers#685 Signed-off-by: Michael Kesper <[email protected]>

rhatdan · 2025-02-17T12:21:00Z

#835 Should fix the --quiet issue.

rhatdan · 2025-02-17T13:14:12Z

@mkesper Rebase and repush and we should get this in before the release.

ericcurtin · 2025-02-17T15:21:06Z

Can we change --dry-run back to --dryrun please

ericcurtin · 2025-02-17T15:22:46Z

If there is inconsistence "--dry-run" -> "--dryrun"

rhatdan · 2025-02-17T16:20:21Z

We should alias the two, since I can never remember to use --dryrun or --dry-run.

Podman actually uses both spellings. podman auto update uses --dry-run and quadlet uses --dryrun.

ericcurtin · 2025-02-17T17:17:13Z

Yeah I actually copied quadlet at the time

mkesper · 2025-02-17T23:01:32Z

I added a clarifying comment to the second dryrun argument. I honestly thought that it was mistakenly added twice.

Add --quiet as mutually exclusive with --debug. Maybe a --verbose switch could also be added. Related: containers#684 Signed-off-by: Michael Kesper <[email protected]>

`version` is already a subcommand. Signed-off-by: Michael Kesper <[email protected]>

Related: containers#684 Signed-off-by: Michael Kesper <[email protected]>

See containers#684 Signed-off-by: Michael Kesper <[email protected]>

sourcery-ai bot reviewed Jan 31, 2025

View reviewed changes

ericcurtin approved these changes Feb 1, 2025

View reviewed changes

jhjaggars reviewed Feb 1, 2025

View reviewed changes

swarajpande5 requested changes Feb 4, 2025

View reviewed changes

mkesper force-pushed the main branch from 94bd95e to 254f04a Compare February 6, 2025 16:56

mkesper requested review from rhatdan, bmahabirbu, maxamillion, cgruver, slp and engelmi as code owners February 6, 2025 16:56

dougsland reviewed Feb 7, 2025

View reviewed changes

mkesper force-pushed the main branch from a979179 to 14e0e5c Compare February 13, 2025 17:15

mkesper changed the title ~~Fix high CPU usage during ramalama pull~~ Unify CLI options (verbosity, version) Feb 13, 2025

mkesper added a commit to mkesper/ramalama that referenced this pull request Feb 13, 2025

Add --quiet option for pulls

3eb633c

Related: containers#685 Signed-off-by: Michael Kesper <[email protected]>

mkesper force-pushed the main branch from 14e0e5c to 3eb633c Compare February 13, 2025 18:22

mkesper added a commit to mkesper/ramalama that referenced this pull request Feb 13, 2025

Add --quiet option for pulls

67c457b

Related: containers#685 Signed-off-by: Michael Kesper <[email protected]>

mkesper force-pushed the main branch from 3eb633c to 67c457b Compare February 13, 2025 21:01

mkesper added a commit to mkesper/ramalama that referenced this pull request Feb 13, 2025

Add --quiet option for pulls

2ec8841

Related: containers#685 Signed-off-by: Michael Kesper <[email protected]>

mkesper force-pushed the main branch from 67c457b to 2ec8841 Compare February 13, 2025 21:28

mkesper added a commit to mkesper/ramalama that referenced this pull request Feb 15, 2025

Add --quiet option for pulls

5ff1475

Related: containers#685 Signed-off-by: Michael Kesper <[email protected]>

mkesper force-pushed the main branch from 2ec8841 to 5ff1475 Compare February 15, 2025 16:32

mkesper added a commit to mkesper/ramalama that referenced this pull request Feb 15, 2025

Add --quiet option for pulls

2ccce44

Related: containers#685 Signed-off-by: Michael Kesper <[email protected]>

mkesper force-pushed the main branch from 5ff1475 to 2ccce44 Compare February 15, 2025 16:38

mkesper force-pushed the main branch from 2ccce44 to 444ac47 Compare February 17, 2025 23:00

mkesper and others added 4 commits February 18, 2025 00:33

Move --quiet flag to common options

565f9ca

Add --quiet as mutually exclusive with --debug. Maybe a --verbose switch could also be added. Related: containers#684 Signed-off-by: Michael Kesper <[email protected]>

Remove --version flag

536efce

`version` is already a subcommand. Signed-off-by: Michael Kesper <[email protected]>

Add --quiet option for pulls

ffbf676

Related: containers#684 Signed-off-by: Michael Kesper <[email protected]>

Clarify second dryrun parser argument

2eab097

See containers#684 Signed-off-by: Michael Kesper <[email protected]>

mkesper force-pushed the main branch from 444ac47 to 2eab097 Compare February 17, 2025 23:33

ericcurtin approved these changes Feb 18, 2025

View reviewed changes

ericcurtin merged commit f7c2302 into containers:main Feb 18, 2025
16 checks passed

		@@ -26,11 +26,11 @@ def pull_config_blob(repos, accept, registry_head, manifest_data):
		download_file(url, config_blob_path, headers=headers, show_progress=False)

	def pull_config_blob(repos, accept, registry_head, manifest_data):
	cfg_hash = manifest_data["config"]["digest"]
	config_blob_path = os.path.join(repos, "blobs", cfg_hash)

	os.makedirs(os.path.dirname(config_blob_path), exist_ok=True)

	url = f"{registry_head}/blobs/{cfg_hash}"
	headers = {"Accept": accept}
	download_file(url, config_blob_path, headers=headers, show_progress=False)

Unify CLI options (verbosity, version) #685

Unify CLI options (verbosity, version) #685

Conversation

mkesper commented Jan 31, 2025 • edited by sourcery-ai bot Loading

Summary by Sourcery

sourcery-ai bot commented Jan 31, 2025 • edited Loading

Reviewer's Guide by Sourcery

Sequence diagram for the modified download process

Class diagram showing modified download components

File-Level Changes

Interacting with Sourcery

Customizing Your Experience

Getting Help

sourcery-ai bot left a comment

Choose a reason for hiding this comment

sourcery-ai bot Jan 31, 2025

Choose a reason for hiding this comment

ericcurtin commented Feb 1, 2025

ericcurtin commented Feb 1, 2025 • edited Loading

Choose a reason for hiding this comment

ericcurtin commented Feb 1, 2025 • edited Loading

ericcurtin commented Feb 1, 2025

ericcurtin commented Feb 1, 2025 • edited Loading

jhjaggars commented Feb 3, 2025

ericcurtin commented Feb 3, 2025 • edited Loading

swarajpande5 Feb 4, 2025 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mkesper commented Feb 5, 2025

mkesper commented Feb 6, 2025

mkesper commented Feb 6, 2025

rhatdan commented Feb 7, 2025

Choose a reason for hiding this comment

ericcurtin Feb 7, 2025 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

dougsland commented Feb 7, 2025

dougsland commented Feb 7, 2025

dougsland commented Feb 7, 2025

rhatdan commented Feb 7, 2025

ericcurtin commented Feb 7, 2025 • edited Loading

ericcurtin commented Feb 7, 2025

rhatdan commented Feb 13, 2025

rhatdan commented Feb 13, 2025

mkesper commented Feb 15, 2025

rhatdan commented Feb 17, 2025

rhatdan commented Feb 17, 2025

ericcurtin commented Feb 17, 2025

ericcurtin commented Feb 17, 2025

rhatdan commented Feb 17, 2025

ericcurtin commented Feb 17, 2025

mkesper commented Feb 17, 2025

mkesper commented Jan 31, 2025 •

edited by sourcery-ai bot

Loading

sourcery-ai bot commented Jan 31, 2025 •

edited

Loading

ericcurtin commented Feb 1, 2025 •

edited

Loading

ericcurtin commented Feb 1, 2025 •

edited

Loading

ericcurtin commented Feb 1, 2025 •

edited

Loading

ericcurtin commented Feb 3, 2025 •

edited

Loading

swarajpande5 Feb 4, 2025 •

edited

Loading

ericcurtin Feb 7, 2025 •

edited

Loading

ericcurtin commented Feb 7, 2025 •

edited

Loading