threads / patch / 57755

patchclone: ignore invalid local refs in remote

Subject: [PATCH] clone: ignore invalid local refs in remote

## tl;dr

6 messages between Apr 18, 2022 and Apr 25, 2022. Diffs are folded; open one to read it.

replies: 5people: 3as markdown or json

Derrick Stolee via GitGitGadget· Apr 18, 2022, 17:12 UTC · lore
From: Derrick Stolee <derrickstolee@github.com>

When cloning directly from a local repository, we load a list of refs based on scanning the $GIT_DIR/refs/ directory of the "server" repository. If files exist in that directory that do not parse as hexadecimal hashes, then the ref array used by write_remote_refs() ends up with some entries with null OIDs. This causes us to hit a BUG() statement in ref_transaction_create():

  BUG: create called without valid new_oid

This BUG() call used to be a die() until 033abf97f (Replace all die("BUG: ...") calls by BUG() ones, 2018-05-02). Before that, the die() was added by f04c5b552 (ref_transaction_create(): check that new_sha1 is valid, 2015-02-17).

The original report for this bug [1] mentioned that this problem did not exist in Git 2.27.0. The failure bisects unsurprisingly to 968f12fda (refs: turn on GIT_REF_PARANOIA by default, 2021-09-24). When GIT_REF_PARANOIA is enabled, this case always fails as far back as I am able to successfully compile and test the Git codebase.

[1] https://github.com/git-for-windows/git/issues/3781

There are two approaches to consider here. One would be to remove this BUG() statement in favor of returning with an error. There are only two callers to ref_transaction_create(), so this would have a limited impact.

The other approach would be to add special casing in 'git clone' to avoid this faulty input to the method.

While I originally started with changing 'git clone', I decided that modifying ref_transaction_create() was a more complete solution. This prevents failing with a BUG() statement when we already have a good way to report an error (including a reason for that error) within the method. Both callers properly check the return value and die() with the error message, so this is an appropriate direction.

The added test helps check against a regression, but does check that our intended error message is handled correctly.

Signed-off-by: Derrick Stolee <derrickstolee@github.com>
---
    clone: ignore invalid local refs in remote
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1214%2Fderrickstolee%2Frefs-bug-v1
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1214/derrickstolee/refs-bug-v1
Pull-Request: https://github.com/gitgitgadget/git/pull/1214
 refs.c                 | 6 ++++--
 t/t5605-clone-local.sh | 9 +++++++++
 2 files changed, 13 insertions(+), 2 deletions(-)
Show changes to 2 files +13 −2

refs.c, t/t5605-clone-local.sh

diff --git a/refs.c b/refs.c
index 1a964505f92..f300f83e4d4 100644
--- a/refs.c
+++ b/refs.c
@@ -1111,8 +1111,10 @@ int ref_transaction_create(struct ref_transaction *transaction,
 			   unsigned int flags, const char *msg,
 			   struct strbuf *err)
 {
-	if (!new_oid || is_null_oid(new_oid))
-		BUG("create called without valid new_oid");
+	if (!new_oid || is_null_oid(new_oid)) {
+		strbuf_addf(err, "'%s' has a null OID", refname);
+		return 1;
+	}
 	return ref_transaction_update(transaction, refname, new_oid,
 				      null_oid(), flags, msg, err);
 }
diff --git a/t/t5605-clone-local.sh b/t/t5605-clone-local.sh
index 7d63365f93a..21ab6192839 100755
--- a/t/t5605-clone-local.sh
+++ b/t/t5605-clone-local.sh
@@ -141,4 +141,13 @@ test_expect_success 'cloning locally respects "-u" for fetching refs' '
 	test_must_fail git clone --bare -u false a should_not_work.git
 '
 
+test_expect_success 'local clone from repo with corrupt refs fails gracefully' '
+	git init corrupt &&
+	test_commit -C corrupt one &&
+	echo a >corrupt/.git/refs/heads/topic &&
+
+	test_must_fail git clone corrupt working 2>err &&
+	grep "has a null OID" err
+'
+
 test_done

base-commit: 11cfe552610386954886543f5de87dcc49ad5735
-- 
gitgitgadget
Ævar Arnfjörð Bjarmason· Apr 18, 2022, 17:30 UTC · re: Derrick Stolee via GitGitGadget · lore

Re: [PATCH] clone: ignore invalid local refs in remote

On Mon, Apr 18 2022, Derrick Stolee via GitGitGadget wrote:
Show 91 quoted lines
> From: Derrick Stolee <derrickstolee@github.com>
>
> When cloning directly from a local repository, we load a list of refs
> based on scanning the $GIT_DIR/refs/ directory of the "server"
> repository. If files exist in that directory that do not parse as
> hexadecimal hashes, then the ref array used by write_remote_refs()
> ends up with some entries with null OIDs. This causes us to hit a BUG()
> statement in ref_transaction_create():
>
>   BUG: create called without valid new_oid
>
> This BUG() call used to be a die() until 033abf97f (Replace all
> die("BUG: ...") calls by BUG() ones, 2018-05-02). Before that, the die()
> was added by f04c5b552 (ref_transaction_create(): check that new_sha1 is
> valid, 2015-02-17).
>
> The original report for this bug [1] mentioned that this problem did not
> exist in Git 2.27.0. The failure bisects unsurprisingly to 968f12fda
> (refs: turn on GIT_REF_PARANOIA by default, 2021-09-24). When
> GIT_REF_PARANOIA is enabled, this case always fails as far back as I am
> able to successfully compile and test the Git codebase.
>
> [1] https://github.com/git-for-windows/git/issues/3781
>
> There are two approaches to consider here. One would be to remove this
> BUG() statement in favor of returning with an error. There are only two
> callers to ref_transaction_create(), so this would have a limited
> impact.
>
> The other approach would be to add special casing in 'git clone' to
> avoid this faulty input to the method.
>
> While I originally started with changing 'git clone', I decided that
> modifying ref_transaction_create() was a more complete solution. This
> prevents failing with a BUG() statement when we already have a good way
> to report an error (including a reason for that error) within the
> method. Both callers properly check the return value and die() with the
> error message, so this is an appropriate direction.
>
> The added test helps check against a regression, but does check that our
> intended error message is handled correctly.
>
> Signed-off-by: Derrick Stolee <derrickstolee@github.com>
> ---
>     clone: ignore invalid local refs in remote
>
> Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1214%2Fderrickstolee%2Frefs-bug-v1
> Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1214/derrickstolee/refs-bug-v1
> Pull-Request: https://github.com/gitgitgadget/git/pull/1214
>
>  refs.c                 | 6 ++++--
>  t/t5605-clone-local.sh | 9 +++++++++
>  2 files changed, 13 insertions(+), 2 deletions(-)
>
> diff --git a/refs.c b/refs.c
> index 1a964505f92..f300f83e4d4 100644
> --- a/refs.c
> +++ b/refs.c
> @@ -1111,8 +1111,10 @@ int ref_transaction_create(struct ref_transaction *transaction,
>  			   unsigned int flags, const char *msg,
>  			   struct strbuf *err)
>  {
> -	if (!new_oid || is_null_oid(new_oid))
> -		BUG("create called without valid new_oid");
> +	if (!new_oid || is_null_oid(new_oid)) {
> +		strbuf_addf(err, "'%s' has a null OID", refname);
> +		return 1;
> +	}
>  	return ref_transaction_update(transaction, refname, new_oid,
>  				      null_oid(), flags, msg, err);
>  }
> diff --git a/t/t5605-clone-local.sh b/t/t5605-clone-local.sh
> index 7d63365f93a..21ab6192839 100755
> --- a/t/t5605-clone-local.sh
> +++ b/t/t5605-clone-local.sh
> @@ -141,4 +141,13 @@ test_expect_success 'cloning locally respects "-u" for fetching refs' '
>  	test_must_fail git clone --bare -u false a should_not_work.git
>  '
>  
> +test_expect_success 'local clone from repo with corrupt refs fails gracefully' '
> +	git init corrupt &&
> +	test_commit -C corrupt one &&
> +	echo a >corrupt/.git/refs/heads/topic &&
> +
> +	test_must_fail git clone corrupt working 2>err &&
> +	grep "has a null OID" err
> +'
> +
>  test_done
>
> base-commit: 11cfe552610386954886543f5de87dcc49ad5735

If I change this test to clone from "$PWD/corrupt" instead we get as far as dying in "not our ref" in upload-pack", after the "client" said "want 0000....".

I don't see anything wrong with this narrow fix, but I wonder how close we are to just doing something useful here for the user. I.e. if we can't parse_object() what we're about to "copy" to e.g. error() on that, but carry on cloning the rest if possible.

We should probably always return a non-zero at the end, suggesting the user omit the bad ref with a refspec. But if we did all that we could usefully recover partially corrupt-repos in this state...

Junio C Hamano· Apr 20, 2022, 20:53 UTC · re: Derrick Stolee via GitGitGadget · lore

Re: [PATCH] clone: ignore invalid local refs in remote

"Derrick Stolee via GitGitGadget" <gitgitgadget@gmail.com> writes:
> From: Derrick Stolee <derrickstolee@github.com>
> Subject: Re: [PATCH] clone: ignore invalid local refs in remote

After seeing the title, I expected that cloning from such a repository with cruft in .git/refs/ directory would issue a warning and succeed without these non-ref files.

But that is not what is happening here?
Show 9 quoted lines
> +test_expect_success 'local clone from repo with corrupt refs fails gracefully' '
> +	git init corrupt &&
> +	test_commit -C corrupt one &&
> +	echo a >corrupt/.git/refs/heads/topic &&
> +
> +	test_must_fail git clone corrupt working 2>err &&
> +	grep "has a null OID" err
> +'
> +
We keep expecting that clone _will_ fail.

So the net change is that we still do not tolerate a corrupt repository and do not let corruption to propagate through cloning, but we diagnose this breakage as an error by calling die(), which is appropriate for dealing with runtime data error, instead of hitting a BUG(), which is reserved for program errors.

I agree with the fixed behaviour and implementation. It just is that "ignore" on the title seems misleading. Other than that, thanks for a good finding and a clean fix.

Show 6 quoted lines
> -	if (!new_oid || is_null_oid(new_oid))
> -		BUG("create called without valid new_oid");
> +	if (!new_oid || is_null_oid(new_oid)) {
> +		strbuf_addf(err, "'%s' has a null OID", refname);
> +		return 1;
> +	}
Derrick Stolee· Apr 21, 2022, 13:29 UTC · re: Junio C Hamano · lore

Re: [PATCH] clone: ignore invalid local refs in remote

On 4/20/2022 4:53 PM, Junio C Hamano wrote:
Show 10 quoted lines
> "Derrick Stolee via GitGitGadget" <gitgitgadget@gmail.com> writes:
> 
>> From: Derrick Stolee <derrickstolee@github.com>
>> Subject: Re: [PATCH] clone: ignore invalid local refs in remote
> 
> After seeing the title, I expected that cloning from such a
> repository with cruft in .git/refs/ directory would issue a warning
> and succeed without these non-ref files.
> 
> But that is not what is happening here?
Sorry, this title of the commit message is stale from a version
where I started making the clones succeed (but without these
bad refs). I changed my mind to only switch BUG() to die() to
avoid giving the impression that we have a "matching" repo after
the clone.
 
Show 21 quoted lines
>> +test_expect_success 'local clone from repo with corrupt refs fails gracefully' '
>> +	git init corrupt &&
>> +	test_commit -C corrupt one &&
>> +	echo a >corrupt/.git/refs/heads/topic &&
>> +
>> +	test_must_fail git clone corrupt working 2>err &&
>> +	grep "has a null OID" err
>> +'
>> +
> 
> We keep expecting that clone _will_ fail.
> 
> So the net change is that we still do not tolerate a corrupt
> repository and do not let corruption to propagate through cloning,
> but we diagnose this breakage as an error by calling die(), which 
> is appropriate for dealing with runtime data error, instead of
> hitting a BUG(), which is reserved for program errors.
> 
> I agree with the fixed behaviour and implementation.  It just is
> that "ignore" on the title seems misleading.  Other than that,
> thanks for a good finding and a clean fix.
Perhaps the commit could instead start with
	clone: die() instead of BUG() on bad refs
?

Thanks, -Stolee

Junio C Hamano· Apr 21, 2022, 18:00 UTC · re: Derrick Stolee · lore

Re: [PATCH] clone: ignore invalid local refs in remote

Derrick Stolee <derrickstolee@github.com> writes:
Show 5 quoted lines
> Sorry, this title of the commit message is stale from a version
> where I started making the clones succeed (but without these
> bad refs). I changed my mind to only switch BUG() to die() to
> avoid giving the impression that we have a "matching" repo after
> the clone.
Ah, that figures ;-)
Show 5 quoted lines
> Perhaps the commit could instead start with
>
> 	clone: die() instead of BUG() on bad refs
>
> ?
Yeah, it would be the title I would have been happy with.
Derrick Stolee via GitGitGadget· Apr 25, 2022, 13:47 UTC · re: Derrick Stolee via GitGitGadget · lore

[PATCH v2] clone: die() instead of BUG() on bad refs

From: Derrick Stolee <derrickstolee@github.com>

When cloning directly from a local repository, we load a list of refs based on scanning the $GIT_DIR/refs/ directory of the "server" repository. If files exist in that directory that do not parse as hexadecimal hashes, then the ref array used by write_remote_refs() ends up with some entries with null OIDs. This causes us to hit a BUG() statement in ref_transaction_create():

  BUG: create called without valid new_oid

This BUG() call used to be a die() until 033abf97f (Replace all die("BUG: ...") calls by BUG() ones, 2018-05-02). Before that, the die() was added by f04c5b552 (ref_transaction_create(): check that new_sha1 is valid, 2015-02-17).

The original report for this bug [1] mentioned that this problem did not exist in Git 2.27.0. The failure bisects unsurprisingly to 968f12fda (refs: turn on GIT_REF_PARANOIA by default, 2021-09-24). When GIT_REF_PARANOIA is enabled, this case always fails as far back as I am able to successfully compile and test the Git codebase.

[1] https://github.com/git-for-windows/git/issues/3781

There are two approaches to consider here. One would be to remove this BUG() statement in favor of returning with an error. There are only two callers to ref_transaction_create(), so this would have a limited impact.

The other approach would be to add special casing in 'git clone' to avoid this faulty input to the method.

While I originally started with changing 'git clone', I decided that modifying ref_transaction_create() was a more complete solution. This prevents failing with a BUG() statement when we already have a good way to report an error (including a reason for that error) within the method. Both callers properly check the return value and die() with the error message, so this is an appropriate direction.

The added test helps check against a regression, but does check that our intended error message is handled correctly.

Signed-off-by: Derrick Stolee <derrickstolee@github.com>
---
    clone: ignore invalid local refs in remote
    
    Update in v2:
    
     * Subject line is improved to better match the message and fix.
     * I did not explore Ævar's idea to optimistically continue the clone as
       much as possible. That seems like a bigger change that could be
       investigated independently.
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1214%2Fderrickstolee%2Frefs-bug-v2
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1214/derrickstolee/refs-bug-v2
Pull-Request: https://github.com/gitgitgadget/git/pull/1214
Range-diff vs v1:
 1:  6c9e85c5a98 ! 1:  9801123c534 clone: ignore invalid local refs in remote
     @@ Metadata
      Author: Derrick Stolee <derrickstolee@github.com>
      
       ## Commit message ##
     -    clone: ignore invalid local refs in remote
     +    clone: die() instead of BUG() on bad refs
      
          When cloning directly from a local repository, we load a list of refs
          based on scanning the $GIT_DIR/refs/ directory of the "server"
 refs.c                 | 6 ++++--
 t/t5605-clone-local.sh | 9 +++++++++
 2 files changed, 13 insertions(+), 2 deletions(-)
Show changes to 2 files +13 −2

refs.c, t/t5605-clone-local.sh

diff --git a/refs.c b/refs.c
index 1a964505f92..f300f83e4d4 100644
--- a/refs.c
+++ b/refs.c
@@ -1111,8 +1111,10 @@ int ref_transaction_create(struct ref_transaction *transaction,
 			   unsigned int flags, const char *msg,
 			   struct strbuf *err)
 {
-	if (!new_oid || is_null_oid(new_oid))
-		BUG("create called without valid new_oid");
+	if (!new_oid || is_null_oid(new_oid)) {
+		strbuf_addf(err, "'%s' has a null OID", refname);
+		return 1;
+	}
 	return ref_transaction_update(transaction, refname, new_oid,
 				      null_oid(), flags, msg, err);
 }
diff --git a/t/t5605-clone-local.sh b/t/t5605-clone-local.sh
index 7d63365f93a..21ab6192839 100755
--- a/t/t5605-clone-local.sh
+++ b/t/t5605-clone-local.sh
@@ -141,4 +141,13 @@ test_expect_success 'cloning locally respects "-u" for fetching refs' '
 	test_must_fail git clone --bare -u false a should_not_work.git
 '
 
+test_expect_success 'local clone from repo with corrupt refs fails gracefully' '
+	git init corrupt &&
+	test_commit -C corrupt one &&
+	echo a >corrupt/.git/refs/heads/topic &&
+
+	test_must_fail git clone corrupt working 2>err &&
+	grep "has a null OID" err
+'
+
 test_done

base-commit: 11cfe552610386954886543f5de87dcc49ad5735
-- 
gitgitgadget

← back to recent threads