Promote x-nbd-server-add-bitmap to stable
v3 cover Promote x-nbd-server-add-bitmap to stable
Or rather, move its functionality into nbd-server-add. And as a side effect, teach qemu-nbd how to export a persistent bitmap without having to go through a qemu process and several QMP commands.
Based-on: <[email protected]>
[jsnow: 0/11 bitmaps: remove x- prefix from QMP api]
Based-on: <[email protected]>
[philmd: qemu-nbd: Rename 'exp' variable clashing with math::exp() symbol]
Available at: https://repo.or.cz/qemu/ericb.git nbd-bitmap-add-v3
Since v2:
- split old patch 1 into 3 parts
- add even more tests of expected error messages
- rebase on top of 'exp' rename
- improve commit messages
- add some R-b where it made sense
001/8:[down] 'nbd: Add some error case testing to iotests 223' 002/8:[down] 'nbd: Forbid nbd-server-stop when server is not running' 003/8:[0011] [FC] 'nbd: Only require disabled bitmap for read-only exports' 004/8:[0020] [FC] 'nbd: Merge nbd_export_set_name into nbd_export_new' 005/8:[0008] [FC] 'nbd: Allow bitmap export during QMP nbd-server-add' 006/8:[----] [--] 'nbd: Remove x-nbd-server-add-bitmap' 007/8:[0009] [FC] 'nbd: Merge nbd_export_bitmap into nbd_export_new' 008/8:[0004] [FC] 'qemu-nbd: Add --bitmap=NAME option'
Eric Blake (8):
nbd: Add some error case testing to iotests 223
nbd: Forbid nbd-server-stop when server is not running
nbd: Only require disabled bitmap for read-only exports
nbd: Merge nbd_export_set_name into nbd_export_new
nbd: Allow bitmap export during QMP nbd-server-add
nbd: Remove x-nbd-server-add-bitmap
nbd: Merge nbd_export_bitmap into nbd_export_new
qemu-nbd: Add --bitmap=NAME option
qemu-nbd.texi | 4 ++ qapi/block.json | 30 ++------ include/block/nbd.h | 12 ++-- blockdev-nbd.c | 36 +++------- hmp.c | 5 +- nbd/server.c | 136 +++++++++++++++++-------------------- qemu-nbd.c | 17 +++-- tests/qemu-iotests/223 | 50 +++++++++++--- tests/qemu-iotests/223.out | 23 +++++-- 9 files changed, 160 insertions(+), 153 deletions(-)
Signature · 1 line
2.20.1
PATCH v3 1/8 nbd: Add some error case testing to iotests 223
Testing success paths is important, but it's also nice to highlight expected failure handling, to show that we don't crash, and so that upcoming tests that change behavior can demonstrate the resulting effects on error paths.
Add the following errors:
Attempting to export without a running server Attempting to start a second server Attempting to export a bad node name Attempting to export a name that is already exported Attempting to export an enabled bitmap Attempting to clean an already cleaned export Attempting to quit server a second time
All of these properly complain except for a second server-stop, which will be fixed next.
---
v3: split out unrelated error path testing and expand [Vladimir]
--- tests/qemu-iotests/223 | 19 +++++++++++++++++-- tests/qemu-iotests/223.out | 7 +++++++ 2 files changed, 24 insertions(+), 2 deletions(-)
tests/qemu-iotests/223 +17 -2
@@ -107,6 +107,7 @@ echo _launch_qemu 2> >(_filter_nbd) +# Intentionally provoke some errors as well, to check error handling silent= _send_qemu_cmd $QEMU_HANDLE '{"execute":"qmp_capabilities"}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"blockdev-add",
@@ -114,17 +115,28 @@ _send_qemu_cmd $QEMU_HANDLE '{"execute":"blockdev-add", "file":{"driver":"file", "filename":"'"$TEST_IMG"'"}}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"block-dirty-bitmap-disable", "arguments":{"node":"n", "name":"b"}}' "return" -_send_qemu_cmd $QEMU_HANDLE '{"execute":"block-dirty-bitmap-disable", - "arguments":{"node":"n", "name":"b2"}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"n"}}' "error" # Attempt add without server _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-start", "arguments":{"addr":{"type":"unix", "data":{"path":"'"$TEST_DIR/nbd"'"}}}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-start", + "arguments":{"addr":{"type":"unix", + "data":{"path":"'"$TEST_DIR/nbd"1'"}}}}' "error" # Attempt second server _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", "arguments":{"device":"n"}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"nosuch"}}' "error" # Attempt to export missing node +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"n"}}' "error" # Attempt to export same name twice _send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", "arguments":{"name":"n", "bitmap":"b"}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", "arguments":{"device":"n", "name":"n2"}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", + "arguments":{"name":"n2", "bitmap":"b2"}}' "error" # Attempt enabled bitmap +_send_qemu_cmd $QEMU_HANDLE '{"execute":"block-dirty-bitmap-disable", + "arguments":{"node":"n", "name":"b2"}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", "arguments":{"name":"n2", "bitmap":"b2"}}' "return"
@@ -157,7 +169,10 @@ _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", "arguments":{"name":"n"}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", "arguments":{"name":"n2"}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", + "arguments":{"name":"n2"}}' "error" # Attempt duplicate clean _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "return" # Oops _send_qemu_cmd $QEMU_HANDLE '{"execute":"quit"}' "return" # success, all done
tests/qemu-iotests/223.out +7 -0
@@ -27,10 +27,15 @@ wrote 2097152/2097152 bytes at offset 2097152 {"return": {}} {"return": {}} {"return": {}} +{"error": {"class": "GenericError", "desc": "NBD server not running"}} {"return": {}} +{"error": {"class": "GenericError", "desc": "NBD server already running"}} {"return": {}} +{"error": {"class": "GenericError", "desc": "Cannot find device=nosuch nor node_name=nosuch"}} +{"error": {"class": "GenericError", "desc": "NBD server already has export named 'n'"}} {"return": {}} {"return": {}} +{"error": {"class": "GenericError", "desc": "Bitmap 'b2' is enabled"}} {"return": {}} {"return": {}}
@@ -62,6 +67,8 @@ read 2097152/2097152 bytes at offset 2097152 {"return": {}} {"return": {}} +{"error": {"class": "GenericError", "desc": "Export 'n2' is not found"}} +{"return": {}} {"return": {}} {"return": {}} *** done
Signature · 1 line
2.20.1
Re: [Qemu-devel] [PATCH v3 1/8] nbd: Add some error case testing to iotests 223
Quoted 5 lines · Eric Blake
Testing success paths is important, but it's also nice to highlight expected failure handling, to show that we don't crash, and so that upcoming tests that change behavior can demonstrate the resulting effects on error paths.
Quoted 7 lines
Add the following errors:
Attempting to export without a running server Attempting to start a second server Attempting to export a bad node name Attempting to export a name that is already exported Attempting to export an enabled bitmap Attempting to clean an already cleaned export
For me "clean" sounds not the same as "remove" (I used to the thought that remove means to remove object from larger object, and clean is clean the object internally (remove its content?), but it continues to exist and all its relations with parents are not touched), but I'm really far from being an expert in English.
Quoted 4 lines
Attempting to quit server a second time
All of these properly complain except for a second server-stop, which will be fixed next.
hm, looks a bit strange to add test first and then fix bug, but on the other hand it's more demonstrative, OK for me.
Quoted 2 lines
Signed-off-by Eric Blake <[email protected]>
Signature · 2 lines
Best regards, Vladimir
Re: [Qemu-devel] [PATCH v3 1/8] nbd: Add some error case testing to iotests 223
Quoted 14 lines · Vladimir Sementsov-Ogievskiy
Quoted 5 lines · Eric Blake
Testing success paths is important, but it's also nice to highlight expected failure handling, to show that we don't crash, and so that upcoming tests that change behavior can demonstrate the resulting effects on error paths.
Quoted 7 lines
Add the following errors:
Attempting to export without a running server Attempting to start a second server Attempting to export a bad node name Attempting to export a name that is already exported Attempting to export an enabled bitmap Attempting to clean an already cleaned export
Quoted 5 lines
For me "clean" sounds not the same as "remove" (I used to the thought that remove means to remove object from larger object, and clean is clean the object internally (remove its content?), but it continues to exist and all its relations with parents are not touched), but I'm really far from being an expert in English.
Sure, I'll s/clean/remove/ in that text.
Quoted 16 lines
Quoted 5 lines
Attempting to quit server a second time
All of these properly complain except for a second server-stop, which will be fixed next.
hm, looks a bit strange to add test first and then fix bug, but on the other hand it's more demonstrative, OK for me.
Quoted 2 lines
Signed-off-by Eric Blake <[email protected]>Reviewed-by Vladimir Sementsov-Ogievskiy <[email protected]>
Signature · 3 lines
Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3226 Virtualization: qemu.org | libvirt.org
PATCH v3 2/8 nbd: Forbid nbd-server-stop when server is not running
In reply to Eric BlakeSince we already forbid other nbd-server commands when not in the right state, it is unlikely that any caller was relying on a second stop to behave as a silent no-op. Update iotest 223 to show the improved behavior.
---
v3: new patch
--- blockdev-nbd.c | 5 +++++ tests/qemu-iotests/223 | 2 +- tests/qemu-iotests/223.out | 2 +- 3 files changed, 7 insertions(+), 2 deletions(-)
blockdev-nbd.c +5 -0
@@ -214,6 +214,11 @@ void qmp_nbd_server_remove(const char *name,void qmp_nbd_server_stop (Error **errp) { +if (!nbd_server) { +error_setg (errp,"NBD server not running" ); +return ; + } +nbd_export_close_all ();nbd_server_free (nbd_server);
tests/qemu-iotests/223 +1 -1
@@ -172,7 +172,7 @@ _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", "arguments":{"name":"n2"}}' "error" # Attempt duplicate clean _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "return" -_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "return" # Oops +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "error" # Again _send_qemu_cmd $QEMU_HANDLE '{"execute":"quit"}' "return" # success, all done
tests/qemu-iotests/223.out +1 -1
@@ -69,6 +69,6 @@ read 2097152/2097152 bytes at offset 2097152 {"return": {}} {"error": {"class": "GenericError", "desc": "Export 'n2' is not found"}} {"return": {}} -{"return": {}} +{"error": {"class": "GenericError", "desc": "NBD server not running"}} {"return": {}} *** done
Signature · 1 line
2.20.1
Re: [Qemu-devel] [PATCH v3 2/8] nbd: Forbid nbd-server-stop when server is not running
Quoted 6 lines · Eric Blake
Since we already forbid other nbd-server commands when not in the right state, it is unlikely that any caller was relying on a second stop to behave as a silent no-op. Update iotest 223 to show the improved behavior.
Signed-off-by Eric Blake <[email protected]>
Quoted 14 lines
---
v3: new patch--- blockdev-nbd.c | 5 +++++ tests/qemu-iotests/223 | 2 +- tests/qemu-iotests/223.out | 2 +- 3 files changed, 7 insertions(+), 2 deletions(-)blockdev-nbd.c +0
-0
blockdev-nbd.c void qmp_nbd_server_remove(const char *name, · line 220
Quoted 7 lines
blockdev-nbd.c +4
-0@@ -214,6 +214,11 @@ void qmp_nbd_server_remove(const char *name,void qmp_nbd_server_stop (Error **errp) { +if (!nbd_server) { +error_setg (errp,"NBD server not running" ); +return ; + }
yes, exactly like in qmp_nbd_server_remove
Quoted 9 lines
blockdev-nbd.c +1
-0+nbd_export_close_all ();nbd_server_free (nbd_server);tests/qemu-iotests/223 +0
-0
tests/qemu-iotests/223 _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", · line 175
Quoted 5 lines
tests/qemu-iotests/223 +1
-1@@ -172,7 +172,7 @@ _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", "arguments":{"name":"n2"}}' "error" # Attempt duplicate clean _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "return" -_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "return" # Oops +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "error" # Again
hm, interesting, but looks like the latest "return" or "error" is an extra thing, we can use "}" for all qmp commands, and it worth adding _send_qmp_cmd wrapper.. But, of course, it's not about these series.
Quoted 16 lines
tests/qemu-iotests/223 +0
-0_send_qemu_cmd $QEMU_HANDLE '{"execute":"quit"}' "return" # success, all donetests/qemu-iotests/223.out +1
-1@@ -69,6 +69,6 @@ read 2097152/2097152 bytes at offset 2097152 {"return": {}} {"error": {"class": "GenericError", "desc": "Export 'n2' is not found"}} {"return": {}} -{"return": {}} +{"error": {"class": "GenericError", "desc": "NBD server not running"}} {"return": {}} *** done
Signature · 2 lines
Best regards, Vladimir
PATCH v3 3/8 nbd: Only require disabled bitmap for read-only exports
In reply to Eric BlakeOur initial implementation of x-nbd-server-add-bitmap put in a restriction because of incremental backups: in that usage, we are exporting one qcow2 file (the temporary overlay target of a blockdev-backup sync:none job) and a dirty bitmap owned by a second qcow2 file (the source of the blockdev-backup, which is the backing file of the temporary). While both qcow2 files are still writable (the target in order to capture copy-on-write of old contents, and the source in order to track live guest writes in the meantime), the NBD client expects to see constant data, including the dirty bitmap. An enabled bitmap in the source would be modified by guest writes, which is at odds with the NBD export being a read-only constant view, hence the initial code choice of enforcing a disabled bitmap (the intent is that the exposed bitmap was disabled in the same transaction that started the blockdev-backup job, although we don't want to track enough state to actually enforce that).
However, consider the case of a bitmap contained in a read-only node (including when the bitmap is found in a backing layer of the active image). Because the node can't be modified, the bitmap won't change due to writes, regardless of whether it is still enabled. Forbidding the export unless the bitmap is disabled is awkward, paritcularly since we can't change the bitmap to be disabled (because the node is read-only).
Alternatively, consider the case of live storage migration, where management directs the destination to create a writable NBD server, then performs a drive-mirror from the source to the mirror, prior to doing the rest of the live migration. Since storage migration can be time-consuming, it may be wise to let the destination include a dirty bitmap to track which portions it has already received, where even if the migration is interrupted and restarted, the source can query the destination block status in order to minimize re-sending data that has not changed in the meantime on a second attempt. Such code has not been written, but it makes sense that letting an active dirty bitmap be exposed and changing alongside writes may prove useful in the future.
Solve both issues by gating the restriction against a disabled bitmap to only happen when the caller has requested a read-only export, and where the BDS that owns the bitmap (whether or not it is the BDS handed to nbd_export_new() or from its backing chain) is still writable.
Update iotest 223 to show the looser behavior by leaving a bitmap enabled the whole run; note that we have to tear down and re-export a node when handling an error.
---
v3: split out unrelated test changes, improve commit message [Vladimir]
v2: new patch
--- nbd/server.c | 7 +++++-- tests/qemu-iotests/223 | 10 +++++++--- tests/qemu-iotests/223.out | 3 ++- 3 files changed, 14 insertions(+), 6 deletions(-)
nbd/server.c +5 -2
@@ -2456,8 +2456,11 @@ void nbd_export_bitmap(NBDExport *exp, const char *bitmap,return ; } -if (bdrv_dirty_bitmap_enabled (bm)) { -error_setg (errp,"Bitmap '%s' is enabled" , bitmap); +if ((exp->nbdflags &NBD_FLAG_READ_ONLY ) &&bdrv_is_writable (bs) && +bdrv_dirty_bitmap_enabled (bm)) { +error_setg (errp, +"Enabled bitmap '%s' incompatible with readonly export" , + bitmap);return ; }
tests/qemu-iotests/223 +7 -3
@@ -61,6 +61,8 @@ echo "=== Create partially sparse image, then add dirty bitmaps ===" echo # Two bitmaps, to contrast granularity issues +# Also note that b will be disabled, while b2 is left enabled, to +# check for read-only interactions _make_test_img -o cluster_size=4k 4M $QEMU_IO -c 'w -P 0x11 1M 2M' "$TEST_IMG" | _filter_qemu_io run_qemu <<EOF
@@ -134,9 +136,11 @@ _send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", "arguments":{"device":"n", "name":"n2"}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", - "arguments":{"name":"n2", "bitmap":"b2"}}' "error" # Attempt enabled bitmap -_send_qemu_cmd $QEMU_HANDLE '{"execute":"block-dirty-bitmap-disable", - "arguments":{"node":"n", "name":"b2"}}' "return" + "arguments":{"name":"n2", "bitmap":"b2"}}' "error" # Enabled vs. read-only +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", + "arguments":{"name":"n2"}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"n", "name":"n2", "writable":true}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", "arguments":{"name":"n2", "bitmap":"b2"}}' "return"
tests/qemu-iotests/223.out +2 -1
@@ -35,7 +35,8 @@ wrote 2097152/2097152 bytes at offset 2097152 {"error": {"class": "GenericError", "desc": "NBD server already has export named 'n'"}} {"return": {}} {"return": {}} -{"error": {"class": "GenericError", "desc": "Bitmap 'b2' is enabled"}} +{"error": {"class": "GenericError", "desc": "Enabled bitmap 'b2' incompatible with readonly export"}} +{"return": {}} {"return": {}} {"return": {}}
Signature · 1 line
2.20.1
Re: [Qemu-devel] [PATCH v3 3/8] nbd: Only require disabled bitmap for read-only exports
Quoted 26 lines · Eric Blake
Our initial implementation of x-nbd-server-add-bitmap put in a restriction because of incremental backups: in that usage, we are exporting one qcow2 file (the temporary overlay target of a blockdev-backup sync:none job) and a dirty bitmap owned by a second qcow2 file (the source of the blockdev-backup, which is the backing file of the temporary). While both qcow2 files are still writable (the target in order to capture copy-on-write of old contents, and the source in order to track live guest writes in the meantime), the NBD client expects to see constant data, including the dirty bitmap. An enabled bitmap in the source would be modified by guest writes, which is at odds with the NBD export being a read-only constant view, hence the initial code choice of enforcing a disabled bitmap (the intent is that the exposed bitmap was disabled in the same transaction that started the blockdev-backup job, although we don't want to track enough state to actually enforce that).
However, consider the case of a bitmap contained in a read-only node (including when the bitmap is found in a backing layer of the active image). Because the node can't be modified, the bitmap won't change due to writes, regardless of whether it is still enabled. Forbidding the export unless the bitmap is disabled is awkward, paritcularly since we can't change the bitmap to be disabled (because the node is read-only).
Quoted 4 lines
Alternatively, consider the case of live storage migration, where management directs the destination to create a writable NBD server, then performs a drive-mirror from the source to the mirror, prior to doing the rest of the live migration.
s/mirror/target ?
Quoted 6 lines
Since storage migration can be time-consuming, it may be wise to let the destination include a dirty bitmap to track which portions it has already received, where even if the migration is interrupted and restarted, the source can query the destination block status in order to minimize re-sending data that has not changed in the meantime on a second attempt.
hmm, dirty bitmap dirtiness don't guarantee that data was written, due to:
1. philosofy: we generally don't care, when something is reported as dirty, when actually it is clean, we assume that it is always safe to consider things to be dirty.
2. granularity: really, at least granularity of target bitmap should correspond to mirror cluster size and target block size, if one of these three things is out of sync, we definitely get [1]
3. errors: bitmaps are set dirty for failed writes too, so [1]
I'm not totally against, I just note that the feature you mean is at least not as simple to implement. And for me it looks like it is more correct to track progress of mirror on source, assuming all success nbd write requests as successful story. At least, as all vm metadata is on source, when storage migration not finished yet.
Quoted 10 lines
Such code has not been written, but it makes sense that letting an active dirty bitmap be exposed and changing alongside writes may prove useful in the future.
Solve both issues by gating the restriction against a disabled bitmap to only happen when the caller has requested a read-only export, and where the BDS that owns the bitmap (whether or not it is the BDS handed to
nbd_export_new()or from its backing chain) is still writable.
Quoted 5 lines
Update iotest 223 to show the looser behavior by leaving a bitmap enabled the whole run; note that we have to tear down and re-export a node when handling an error.
Signed-off-by Eric Blake <[email protected]>
anyway, I'm OK with new if-condition, and for me something like
No real reasons to forbid export of enabled bitmap, user should understand, that block-status result may not correspond to reality if bitmap may be changed by guest-write or something in parallel (including client's self requests). So, it should be OK to drop the restriction at all, but let's keep a smaller one to prevent dangerous wrong management error..
would be enough description so,
Or, we can allow exporting enabled bitmaps only for read-only nodes, until we don't have real reasoning for your second case.
Quoted 70 lines
---
v3: split out unrelated test changes, improve commit message [Vladimir]
v2: new patch--- nbd/server.c | 7 +++++-- tests/qemu-iotests/223 | 10 +++++++--- tests/qemu-iotests/223.out | 3 ++- 3 files changed, 14 insertions(+), 6 deletions(-)nbd/server.c +5
-2@@ -2456,8 +2456,11 @@ void nbd_export_bitmap(NBDExport *exp, const char *bitmap,return ; } -if (bdrv_dirty_bitmap_enabled (bm)) { -error_setg (errp,"Bitmap '%s' is enabled" , bitmap); +if ((exp->nbdflags &NBD_FLAG_READ_ONLY ) &&bdrv_is_writable (bs) && +bdrv_dirty_bitmap_enabled (bm)) { +error_setg (errp, +"Enabled bitmap '%s' incompatible with readonly export" , + bitmap);return ; }tests/qemu-iotests/223 +7
-3@@ -61,6 +61,8 @@ echo "=== Create partially sparse image, then add dirty bitmaps ===" echo # Two bitmaps, to contrast granularity issues +# Also note that b will be disabled, while b2 is left enabled, to +# check for read-only interactions _make_test_img -o cluster_size=4k 4M $QEMU_IO -c 'w -P 0x11 1M 2M' "$TEST_IMG" | _filter_qemu_io run_qemu <<EOF@@ -134,9 +136,11 @@ _send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", "arguments":{"device":"n", "name":"n2"}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", - "arguments":{"name":"n2", "bitmap":"b2"}}' "error" # Attempt enabled bitmap -_send_qemu_cmd $QEMU_HANDLE '{"execute":"block-dirty-bitmap-disable", - "arguments":{"node":"n", "name":"b2"}}' "return" + "arguments":{"name":"n2", "bitmap":"b2"}}' "error" # Enabled vs. read-only +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", + "arguments":{"name":"n2"}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"n", "name":"n2", "writable":true}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", "arguments":{"name":"n2", "bitmap":"b2"}}' "return"tests/qemu-iotests/223.out +2
-1@@ -35,7 +35,8 @@ wrote 2097152/2097152 bytes at offset 2097152 {"error": {"class": "GenericError", "desc": "NBD server already has export named 'n'"}} {"return": {}} {"return": {}} -{"error": {"class": "GenericError", "desc": "Bitmap 'b2' is enabled"}} +{"error": {"class": "GenericError", "desc": "Enabled bitmap 'b2' incompatible with readonly export"}} +{"return": {}} {"return": {}} {"return": {}}
Signature · 2 lines
Best regards, Vladimir
Re: [Qemu-devel] [PATCH v3 3/8] nbd: Only require disabled bitmap for read-only exports
Quoted 27 lines · Vladimir Sementsov-Ogievskiy
Quoted 26 lines · Eric Blake
Our initial implementation of x-nbd-server-add-bitmap put in a restriction because of incremental backups: in that usage, we are exporting one qcow2 file (the temporary overlay target of a blockdev-backup sync:none job) and a dirty bitmap owned by a second qcow2 file (the source of the blockdev-backup, which is the backing file of the temporary). While both qcow2 files are still writable (the target in order to capture copy-on-write of old contents, and the source in order to track live guest writes in the meantime), the NBD client expects to see constant data, including the dirty bitmap. An enabled bitmap in the source would be modified by guest writes, which is at odds with the NBD export being a read-only constant view, hence the initial code choice of enforcing a disabled bitmap (the intent is that the exposed bitmap was disabled in the same transaction that started the blockdev-backup job, although we don't want to track enough state to actually enforce that).
However, consider the case of a bitmap contained in a read-only node (including when the bitmap is found in a backing layer of the active image). Because the node can't be modified, the bitmap won't change due to writes, regardless of whether it is still enabled. Forbidding the export unless the bitmap is disabled is awkward, paritcularly since we can't change the bitmap to be disabled (because the node is read-only).
Quoted 6 lines
Quoted 4 lines
Alternatively, consider the case of live storage migration, where management directs the destination to create a writable NBD server, then performs a drive-mirror from the source to the mirror, prior to doing the rest of the live migration.
s/mirror/target ?
Yes.
Quoted 15 lines
Quoted 7 lines
Since storage migration can be time-consuming, it may be wise to let the destination include a dirty bitmap to track which portions it has already received, where even if the migration is interrupted and restarted, the source can query the destination block status in order to minimize re-sending data that has not changed in the meantime on a second attempt.
hmm, dirty bitmap dirtiness don't guarantee that data was written, due to:
1. philosofy: we generally don't care, when something is reported as dirty, when actually it is clean, we assume that it is always safe to consider things to be dirty.
Quoted 5 lines
2. granularity: really, at least granularity of target bitmap should correspond to mirror cluster size and target block size, if one of these three things is out of sync, we definitely get [1]
3. errors: bitmaps are set dirty for failed writes too, so [1]
Good points.
Quoted 6 lines
I'm not totally against, I just note that the feature you mean is at least not as simple to implement. And for me it looks like it is more correct to track progress of mirror on source, assuming all success nbd write requests as successful story. At least, as all vm metadata is on source, when storage migration not finished yet.
I'm fine adding a caveat that while allowing reads from a modifying dirty bitmap may help, they may not be trivial to use correctly.
Quoted 4 lines
Quoted 4 lines
Such code has not been written, but it makes sense that letting an active dirty bitmap be exposed and changing alongside writes may prove useful in the future.
to go along with my caveat that such code does not yet exist anyways :)
Quoted 15 lines
Quoted 7 lines
Solve both issues by gating the restriction against a disabled bitmap to only happen when the caller has requested a read-only export, and where the BDS that owns the bitmap (whether or not it is the BDS handed to
nbd_export_new()or from its backing chain) is still writable.Quoted 5 lines
Update iotest 223 to show the looser behavior by leaving a bitmap enabled the whole run; note that we have to tear down and re-export a node when handling an error.
Signed-off-by Eric Blake <[email protected]>anyway, I'm OK with new if-condition, and for me something like
Quoted 8 lines
No real reasons to forbid export of enabled bitmap, user should understand, that block-status result may not correspond to reality if bitmap may be changed by guest-write or something in parallel (including client's self requests). So, it should be OK to drop the restriction at all, but let's keep a smaller one to prevent dangerous wrong management error..
would be enough description so,
Reviewed-by Vladimir Sementsov-Ogievskiy <[email protected]>
Okay. Thanks for the review!
Signature · 3 lines
Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3226 Virtualization: qemu.org | libvirt.org
PATCH v3 4/8 nbd: Merge nbd_export_set_name into nbd_export_new
In reply to Eric BlakeThe existing NBD code had a weird split where nbd_export_new() created an export but did not add it to the list of exported names until a later nbd_export_set_name() came along and grabbed a second reference on the object; later, the first call to nbd_export_close() drops the second reference while removing the export from the list. This is in part because the QAPI NbdServerRemoveNode enum documents the possibility of adding a mode where we could do a soft disconnect: preventing new clients, but waiting for existing clients to gracefully quit, based on the mode used when calling nbd_export_close().
But in spite of all that, note that we never change the name of an NBD export while it is exposed, which means it is easier to just inline the process of setting the name as part of creating the export.
Inline the contents of nbd_export_set_name() and nbd_export_set_description() into the two points in an export lifecycle where they matter, then adjust both callers to pass the name up front. Note that for creation, all callers pass a non-NULL name, (passing NULL at creation was for old style servers, but we removed support for that in commit 7f7dfe2a), so we can add an assert and do things unconditionally; but for cleanup, because of the dual nature of nbd_export_close(), we still have to be careful to avoid use-after-free. Along the way, add a comment reminding ourselves of the potential of adding a middle mode disconnect.
---
v3: add comment on potential close mode semantics [Vladimir], rebase to 'exp' rename [Phillipe]; no major code change so R-b added
--- include/block/nbd.h | 3 +-- blockdev-nbd.c | 5 ++--- nbd/server.c | 52 ++++++++++++++++++++------------------------- qemu-nbd.c | 8 +++---- 4 files changed, 29 insertions(+), 39 deletions(-)
include/block/nbd.h +1 -2
@@ -295,6 +295,7 @@ typedef struct NBDExport NBDExport;typedef struct NBDClient NBDClient ;NBDExport *nbd_export_new (BlockDriverState *bs,off_t dev_offset,off_t size, +const char *name,const char *description,uint16_t nbdflags,void (*close)(NBDExport *),bool writethrough,BlockBackend *on_eject_blk,Error **errp);
@@ -306,8 +307,6 @@ void nbd_export_put(NBDExport *exp);BlockBackend *nbd_export_get_blockdev (NBDExport *exp);NBDExport *nbd_export_find (const char *name); -void nbd_export_set_name (NBDExport *exp,const char *name); -void nbd_export_set_description (NBDExport *exp,const char *description);void nbd_export_close_all (void ); void nbd_client_new(QIOChannelSocket *sioc,
blockdev-nbd.c +2 -3
@@ -174,14 +174,13 @@ void qmp_nbd_server_add(const char *device, bool has_name, const char *name, writable = false; } - exp =nbd_export_new (bs,0 ,-1 , writable ?0 :NBD_FLAG_READ_ONLY , + exp =nbd_export_new (bs,0 ,-1 , name,NULL , + writable ?0 :NBD_FLAG_READ_ONLY ,NULL , false, on_eject_blk, errp);if (!exp) {return ; } -nbd_export_set_name (exp, name); -/* The list of named exports has a strong reference to this export now and * our only way of accessing it is through nbd_export_find(), so we can drop * the strong reference that is @exp. */
nbd/server.c +23 -29
94 lines not on this page · raw message
qemu-nbd.c +3 -5
@@ -1015,11 +1015,9 @@ int main(int argc, char **argv) } } - export =nbd_export_new (bs, dev_offset, fd_size, nbdflags, - nbd_export_closed, writethrough, -NULL , &error_fatal); -nbd_export_set_name (export, export_name); -nbd_export_set_description (export, export_description); + export =nbd_export_new (bs, dev_offset, fd_size, export_name, + export_description, nbdflags, nbd_export_closed, + writethrough,NULL , &error_fatal);if (device) {#if HAVE_NBD_DEVICE
Signature · 1 line
2.20.1
PATCH v3 5/8 nbd: Allow bitmap export during QMP nbd-server-add
In reply to Eric BlakeWith the experimental x-nbd-server-add-bitmap command, there was a window of time where an NBD client could see the export but not the associated dirty bitmap, which can cause a client that planned on using the dirty bitmap to be forced to treat the entire image as dirty as a safety fallback. Furthermore, if the QMP client successfully exports a disk but then fails to add the bitmap, it has to take on the burden of removing the export. Since we don't allow changing the exposed dirty bitmap (whether to a different bitmap, or removing advertisement of the bitmap), it is nicer to make the bitmap tied to the export at the time the export is created, with automatic failure to export if the bitmap is not available.
The experimental command included an optional 'bitmap-export-name' field for remapping the name exposed over NBD to be different from the bitmap name stored on disk. However, my libvirt demo code for implementing differential backups on top of persistent bitmaps did not need to take advantage of that feature (it is instead possible to create a new temporary bitmap with the desired name, use block-dirty-bitmap-merge to merge one or more persistent bitmaps into the temporary, then associate the temporary with the NBD export, if control is needed over the exported bitmap name). Hence, I'm not copying that part of the experiment over to the stable addition. For more details on the libvirt demo, see
https://www.redhat.com/archives/libvir-list/2018-October/msg01254.html,
https://kvmforum2018.sched.com/event/FzuB/facilitating-incremental-backup-eric-blake-red-hat
This patch focuses on the user interface, and reduces (but does not completely eliminate) the window where an NBD client can see the export but not the dirty bitmap, with less work to clean up after errors. Later patches will add further cleanups now that this interface is declared stable via a single QMP command, including removing the race window.
Update test 223 to use the new interface.
--- qapi/block.json | 7 ++++++- blockdev-nbd.c | 12 +++++++++++- hmp.c | 5 +++-- tests/qemu-iotests/223 | 19 ++++++++----------- tests/qemu-iotests/223.out | 5 +---- 5 files changed, 29 insertions(+), 19 deletions(-)
qapi/block.json +6 -1
@@ -246,6 +246,10 @@ # # @writable: Whether clients should be able to write to the device via the # NBD connection (default false). + +# @bitmap: Also export the dirty bitmap reachable from @device, so the +# NBD client can use NBD_OPT_SET_META_CONTEXT with +# "qemu:dirty-bitmap:NAME" to inspect the bitmap. (since 4.0) # # Returns: error if the server is not running, or export with the same name # already exists.
@@ -253,7 +257,8 @@ # Since: 1.3.0 ## { 'command': 'nbd-server-add', - 'data': {'device': 'str', '*name': 'str', '*writable': 'bool'} } + 'data': {'device': 'str', '*name': 'str', '*writable': 'bool', + '*bitmap': 'str' } } ## # @NbdServerRemoveMode:
blockdev-nbd.c +11 -1
@@ -140,7 +140,8 @@ void qmp_nbd_server_start(SocketAddressLegacy *addr, }void qmp_nbd_server_add (const char *device,bool has_name,const char *name, -bool has_writable,bool writable,Error **errp) +bool has_writable,bool writable, +bool has_bitmap,const char *bitmap,Error **errp) {BlockDriverState *bs =NULL ;BlockBackend *on_eject_blk;
@@ -185,6 +186,15 @@ void qmp_nbd_server_add(const char *device, bool has_name, const char *name,* our only way of accessing it is through nbd_export_find(), so we can drop * the strong reference that is @exp. */ nbd_export_put (exp); + +if (has_bitmap) { +Error *err =NULL ; +nbd_export_bitmap (exp, bitmap, bitmap, &err); +if (err) { +error_propagate (errp, err); +nbd_export_remove (exp,NBD_SERVER_REMOVE_MODE_HARD ,NULL ); + } + } } void qmp_nbd_server_remove(const char *name,
hmp.c +3 -2
@@ -2326,7 +2326,7 @@ void hmp_nbd_server_start(Monitor *mon, const QDict *qdict) }qmp_nbd_server_add (info->value->device, false,NULL , - true, writable, &local_err); + true, writable, false,NULL , &local_err);if (local_err !=NULL ) {qmp_nbd_server_stop (NULL );
@@ -2347,7 +2347,8 @@ void hmp_nbd_server_add(Monitor *mon, const QDict *qdict)bool writable =qdict_get_try_bool (qdict,"writable" , false);Error *local_err =NULL ; -qmp_nbd_server_add (device, !!name, name, true, writable, &local_err); +qmp_nbd_server_add (device, !!name, name, true, writable, + false,NULL , &local_err);hmp_handle_error (mon, &local_err); }
tests/qemu-iotests/223 +8 -11
@@ -126,23 +126,20 @@ _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-start", "arguments":{"addr":{"type":"unix", "data":{"path":"'"$TEST_DIR/nbd"1'"}}}}' "error" # Attempt second server _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", - "arguments":{"device":"n"}}' "return" + "arguments":{"device":"n", "bitmap":"b"}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", "arguments":{"device":"nosuch"}}' "error" # Attempt to export missing node _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", "arguments":{"device":"n"}}' "error" # Attempt to export same name twice -_send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", - "arguments":{"name":"n", "bitmap":"b"}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", - "arguments":{"device":"n", "name":"n2"}}' "return" -_send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", - "arguments":{"name":"n2", "bitmap":"b2"}}' "error" # Enabled vs. read-only -_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", - "arguments":{"name":"n2"}}' "return" + "arguments":{"device":"n", "name":"n2", + "bitmap":"b2"}}' "error" # enabled vs. read-only _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", - "arguments":{"device":"n", "name":"n2", "writable":true}}' "return" -_send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", - "arguments":{"name":"n2", "bitmap":"b2"}}' "return" + "arguments":{"device":"n", "name":"n2", + "bitmap":"b3"}}' "error" # Missing bitmap +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"n", "name":"n2", "writable":true, + "bitmap":"b2"}}' "return" echo echo "=== Contrast normal status to large granularity dirty-bitmap ==="
tests/qemu-iotests/223.out +1 -4
@@ -33,11 +33,8 @@ wrote 2097152/2097152 bytes at offset 2097152 {"return": {}} {"error": {"class": "GenericError", "desc": "Cannot find device=nosuch nor node_name=nosuch"}} {"error": {"class": "GenericError", "desc": "NBD server already has export named 'n'"}} -{"return": {}} -{"return": {}} {"error": {"class": "GenericError", "desc": "Enabled bitmap 'b2' incompatible with readonly export"}} -{"return": {}} -{"return": {}} +{"error": {"class": "GenericError", "desc": "Bitmap 'b3' is not found"}} {"return": {}} === Contrast normal status to large granularity dirty-bitmap ===
Signature · 1 line
2.20.1
PATCH v3 6/8 nbd: Remove x-nbd-server-add-bitmap
In reply to Eric BlakeNow that nbd-server-add can do the same functionality (well, other than making the exported bitmap name different than the underlying bitamp - but we argued that was not essential, since it is just as easy to create a new non-persistent bitmap with the desired name), we no longer need the experimental separate command.
--- qapi/block.json | 23 ----------------------- blockdev-nbd.c | 23 ----------------------- 2 files changed, 46 deletions(-)
qapi/block.json +0 -23
@@ -301,29 +301,6 @@ { 'command': 'nbd-server-remove', 'data': {'name': 'str', '*mode': 'NbdServerRemoveMode'} } -## -# @x-nbd-server-add-bitmap: -# -# Expose a dirty bitmap associated with the selected export. The bitmap search -# starts at the device attached to the export, and includes all backing files. -# The exported bitmap is then locked until the NBD export is removed. -# -# @name: Export name. -# -# @bitmap: Bitmap name to search for. -# -# @bitmap-export-name: How the bitmap will be seen by nbd clients -# (default @bitmap) -# -# Note: the client must use NBD_OPT_SET_META_CONTEXT with a query of -# "qemu:dirty-bitmap:NAME" (where NAME matches @bitmap-export-name) to access -# the exposed bitmap. -# -# Since: 3.0 -## - { 'command': 'x-nbd-server-add-bitmap', - 'data': {'name': 'str', 'bitmap': 'str', '*bitmap-export-name': 'str'} } - ## # @nbd-server-stop: #
blockdev-nbd.c +0 -23
@@ -233,26 +233,3 @@ void qmp_nbd_server_stop(Error **errp)nbd_server_free (nbd_server); nbd_server =NULL ; } - -void qmp_x_nbd_server_add_bitmap (const char *name,const char *bitmap, -bool has_bitmap_export_name, -const char *bitmap_export_name, -Error **errp) -{ -NBDExport *exp; - -if (!nbd_server) { -error_setg (errp,"NBD server not running" ); -return ; - } - - exp =nbd_export_find (name); -if (exp ==NULL ) { -error_setg (errp,"Export '%s' is not found" , name); -return ; - } - -nbd_export_bitmap (exp, bitmap, - has_bitmap_export_name ? bitmap_export_name : bitmap, - errp); -}
Signature · 1 line
2.20.1
PATCH v3 7/8 nbd: Merge nbd_export_bitmap into nbd_export_new
In reply to Eric BlakeWe only have one caller that wants to export a bitmap name, which it does right after creation of the export. But there is still a brief window of time where an NBD client could see the export but not the dirty bitmap, which a robust client would have to interpret as meaning the entire image should be treated as dirty. Better is to eliminate the window entirely, by inlining nbd_export_bitmap() into nbd_export_new(), and refusing to create the bitmap in the first place if the requested bitmap can't be located.
We also no longer need logic for setting a different bitmap name compared to the bitmap being exported.
---
v3: rebase to 'exp' rename [Phillipe]
--- include/block/nbd.h | 9 ++--- blockdev-nbd.c | 11 +----- nbd/server.c | 87 +++++++++++++++++++++------------------------ qemu-nbd.c | 5 +-- 4 files changed, 47 insertions(+), 65 deletions(-)
include/block/nbd.h +3 -6
@@ -296,9 +296,9 @@ typedef struct NBDClient NBDClient;NBDExport *nbd_export_new (BlockDriverState *bs,off_t dev_offset,off_t size,const char *name,const char *description, -uint16_t nbdflags,void (*close)(NBDExport *), -bool writethrough,BlockBackend *on_eject_blk, -Error **errp); +const char *bitmap,uint16_t nbdflags, +void (*close)(NBDExport *),bool writethrough, +BlockBackend *on_eject_blk,Error **errp);void nbd_export_close (NBDExport *exp);void nbd_export_remove (NBDExport *exp,NbdServerRemoveMode mode,Error **errp);void nbd_export_get (NBDExport *exp);
@@ -319,9 +319,6 @@ void nbd_client_put(NBDClient *client);void nbd_server_start (SocketAddress *addr,const char *tls_creds,Error **errp); -void nbd_export_bitmap (NBDExport *exp,const char *bitmap, -const char *bitmap_export_name,Error **errp); -/* nbd_read * Reads @size bytes from @ioc. Returns 0 on success. */
blockdev-nbd.c +1 -10
@@ -175,7 +175,7 @@ void qmp_nbd_server_add(const char *device, bool has_name, const char *name, writable = false; } - exp =nbd_export_new (bs,0 ,-1 , name,NULL , + exp =nbd_export_new (bs,0 ,-1 , name,NULL , bitmap, writable ?0 :NBD_FLAG_READ_ONLY ,NULL , false, on_eject_blk, errp);if (!exp) {
@@ -186,15 +186,6 @@ void qmp_nbd_server_add(const char *device, bool has_name, const char *name,* our only way of accessing it is through nbd_export_find(), so we can drop * the strong reference that is @exp. */ nbd_export_put (exp); - -if (has_bitmap) { -Error *err =NULL ; -nbd_export_bitmap (exp, bitmap, bitmap, &err); -if (err) { -error_propagate (errp, err); -nbd_export_remove (exp,NBD_SERVER_REMOVE_MODE_HARD ,NULL ); - } - } } void qmp_nbd_server_remove(const char *name,
nbd/server.c +40 -47
102 lines not on this page · raw message
qemu-nbd.c +3 -2
@@ -1016,8 +1016,9 @@ int main(int argc, char **argv) } export =nbd_export_new (bs, dev_offset, fd_size, export_name, - export_description, nbdflags, nbd_export_closed, - writethrough,NULL , &error_fatal); + export_description,NULL , nbdflags, + nbd_export_closed, writethrough,NULL , + &error_fatal);if (device) {#if HAVE_NBD_DEVICE
Signature · 1 line
2.20.1
PATCH v3 8/8 qemu-nbd: Add --bitmap=NAME option
In reply to Eric BlakeHaving to fire up qemu, then use QMP commands for nbd-server-start and nbd-server-add, just to expose a persistent dirty bitmap, is rather tedious. Make it possible to expose a dirty bitmap using just qemu-nbd (of course, for now this only works when qemu-nbd is visiting a BDS formatted as qcow2).
Of course, any good feature also needs unit testing, so expand iotest 223 to cover it.
---
v3: rebase to earlier changes
--- qemu-nbd.texi | 4 ++++ qemu-nbd.c | 10 ++++++++-- tests/qemu-iotests/223 | 18 +++++++++++++++++- tests/qemu-iotests/223.out | 12 +++++++++++- 4 files changed, 40 insertions(+), 4 deletions(-)
qemu-nbd.texi +4 -0
@@ -45,6 +45,10 @@ auto-detecting Export the disk as read-only @item -P, --partition=@var{num} Only expose partition @var{num} +@item -B, --bitmap=@var{name} +If @var{filename} has a qcow2 persistent bitmap @var{name}, expose +that bitmap via the ``qemu:dirty-bitmap:@var{name}'' context +accessible through NBD_OPT_SET_META_CONTEXT. @item -s, --snapshot Use @var{filename} as an external snapshot, create a temporary file with backing_file=@var{filename}, redirect the write to
qemu-nbd.c +8 -2
@@ -95,6 +95,7 @@ static void usage(const char *name)"Exposing part of the image:\n" " -o, --offset=OFFSET offset into the image\n" " -P, --partition=NUM only expose partition NUM\n" +" -B, --bitmap=NAME expose a persistent dirty bitmap\n" "\n" "General purpose options:\n" " --object type,id=ID,... define an object such as 'secret' for providing\n"
@@ -509,7 +510,7 @@ int main(int argc, char **argv)off_t fd_size;QemuOpts *sn_opts =NULL ;const char *sn_id_or_name =NULL ; -const char *sopt ="hVb:o:p:rsnP:c:dvk:e:f:tl:x:T:D:" ; +const char *sopt ="hVb:o:p:rsnP:c:dvk:e:f:tl:x:T:D:B:" ;struct option lopt[] = { {"help" , no_argument,NULL ,'h' }, {"version" , no_argument,NULL ,'V' },
@@ -519,6 +520,7 @@ int main(int argc, char **argv) { "offset", required_argument, NULL, 'o' }, { "read-only", no_argument, NULL, 'r' }, { "partition", required_argument, NULL, 'P' }, + { "bitmap", required_argument, NULL, 'B' }, { "connect", required_argument, NULL, 'c' }, { "disconnect", no_argument, NULL, 'd' }, { "snapshot", no_argument, NULL, 's' },
@@ -558,6 +560,7 @@ int main(int argc, char **argv)QDict *options =NULL ;const char *export_name ="" ;/* Default export name */ const char *export_description =NULL ; +const char *bitmap =NULL ;const char *tlscredsid =NULL ;bool imageOpts = false;bool writethrough = true;
@@ -695,6 +698,9 @@ int main(int argc, char **argv)exit (EXIT_FAILURE ); }break ; +case 'B' : + bitmap = optarg; +break ;case 'k' : sockpath = optarg;if (sockpath[0 ] !='/' ) {
@@ -1016,7 +1022,7 @@ int main(int argc, char **argv) } export =nbd_export_new (bs, dev_offset, fd_size, export_name, - export_description,NULL , nbdflags, + export_description, bitmap, nbdflags, nbd_export_closed, writethrough,NULL , &error_fatal);
tests/qemu-iotests/223 +17 -1
@@ -25,6 +25,7 @@ status=1 # failure is the default! _cleanup() { + nbd_server_stop _cleanup_test_img _cleanup_qemu rm -f "$TEST_DIR/nbd"
@@ -35,6 +36,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15 . ./common.rc . ./common.filter . ./common.qemu +. ./common.nbd _supported_fmt qcow2 _supported_proto file # uses NBD as well
@@ -163,7 +165,7 @@ $QEMU_IMG map --output=json --image-opts \ "$IMG,x-dirty-bitmap=qemu:dirty-bitmap:b2" | _filter_qemu_img_map echo -echo "=== End NBD server ===" +echo "=== End qemu NBD server ===" echo _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove",
@@ -176,6 +178,20 @@ _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "error" # Again _send_qemu_cmd $QEMU_HANDLE '{"execute":"quit"}' "return" +echo +echo "=== Use qemu-nbd as server ===" +echo + +nbd_server_start_unix_socket -r -f $IMGFMT -B b "$TEST_IMG" +IMG="driver=nbd,server.type=unix,server.path=$nbd_unix_socket" +$QEMU_IMG map --output=json --image-opts \ + "$IMG,x-dirty-bitmap=qemu:dirty-bitmap:b" | _filter_qemu_img_map + +nbd_server_start_unix_socket -f $IMGFMT -B b2 "$TEST_IMG" +IMG="driver=nbd,server.type=unix,server.path=$nbd_unix_socket" +$QEMU_IMG map --output=json --image-opts \ + "$IMG,x-dirty-bitmap=qemu:dirty-bitmap:b2" | _filter_qemu_img_map + # success, all done echo '*** done' rm -f $seq.full
tests/qemu-iotests/223.out +11 -1
@@ -61,7 +61,7 @@ read 2097152/2097152 bytes at offset 2097152 { "start": 1024, "length": 2096128, "depth": 0, "zero": false, "data": true}, { "start": 2097152, "length": 2097152, "depth": 0, "zero": false, "data": false}] -=== End NBD server === +=== End qemu NBD server === {"return": {}} {"return": {}}
@@ -69,4 +69,14 @@ read 2097152/2097152 bytes at offset 2097152 {"return": {}} {"error": {"class": "GenericError", "desc": "NBD server not running"}} {"return": {}} + +=== Use qemu-nbd as server === + +[{ "start": 0, "length": 65536, "depth": 0, "zero": false, "data": false}, +{ "start": 65536, "length": 2031616, "depth": 0, "zero": false, "data": true}, +{ "start": 2097152, "length": 2097152, "depth": 0, "zero": false, "data": false}] +[{ "start": 0, "length": 512, "depth": 0, "zero": false, "data": true}, +{ "start": 512, "length": 512, "depth": 0, "zero": false, "data": false}, +{ "start": 1024, "length": 2096128, "depth": 0, "zero": false, "data": true}, +{ "start": 2097152, "length": 2097152, "depth": 0, "zero": false, "data": false}] *** done
Signature · 1 line
2.20.1
Quoted 19 lines · Eric Blake
Or rather, move its functionality into nbd-server-add. And as a side effect, teach qemu-nbd how to export a persistent bitmap without having to go through a qemu process and several QMP commands.
Based-on: <[email protected]>
[jsnow: 0/11 bitmaps: remove x- prefix from QMP api]
Based-on: <[email protected]>
[philmd: qemu-nbd: Rename 'exp' variable clashing withmath::exp()symbol]Available at: https://repo.or.cz/qemu/ericb.git nbd-bitmap-add-v3
Since v2:
- split old patch 1 into 3 parts
- add even more tests of expected error messages
- rebase on top of 'exp' rename
- improve commit messages
- add some R-b where it made sense
Quoted 8 lines
001/8:[down] 'nbd: Add some error case testing to iotests 223' 002/8:[down] 'nbd: Forbid nbd-server-stop when server is not running' 003/8:[0011] [FC] 'nbd: Only require disabled bitmap for read-only exports' 004/8:[0020] [FC] 'nbd: Merge
nbd_export_set_nameintonbd_export_new' 005/8:[0008] [FC] 'nbd: Allow bitmap export during QMP nbd-server-add' 006/8:[----] [--] 'nbd: Remove x-nbd-server-add-bitmap' 007/8:[0009] [FC] 'nbd: Mergenbd_export_bitmapintonbd_export_new' 008/8:[0004] [FC] 'qemu-nbd: Add --bitmap=NAME option'
Queuing on my NBD tree; I'm planning on sending a pull request Monday morning that includes these patches, unless a last-minute review finding changes things.
https://repo.or.cz/qemu/ericb.git nbd
Signature · 3 lines
Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3226 Virtualization: qemu.org | libvirt.org