block_hotplug_with_max_map_count: creates new test case #4225

mcasquer · 2024-12-02T14:01:40Z

block_hotplug_with_max_map_count: creates new test case

Creates new test case that boots up a VM with a small max_map_count,
then hotplugs 15 disks and checks the VM status.

Signed-off-by: mcasquer [email protected]
ID: 3130

mcasquer · 2024-12-02T14:02:10Z

 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: STARTED
 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: PASS (61.24 s)
RESULTS    : PASS 1 | ERROR 0 | FAIL 0 | SKIP 0 | WARN 0 | INTERRUPT 0 | CANCEL 0

mcasquer · 2024-12-02T14:06:22Z

@qingwangrh please could you review this PR? Thanks!

mcasquer · 2024-12-10T09:42:46Z

This is a kindly reminder, @qingwangrh please could you review this PR? Thanks !

qingwangrh · 2024-12-10T12:11:24Z

qemu/tests/cfg/block_hotplug_with_max_map_count.cfg

+    maxmem_mem = 80G
+    mem = 4096
+    pcie_extra_root_port = 30
+    images += " stg1 stg2 stg3 stg4 stg5 stg6 stg7 stg8 stg9 stg10 stg11 stg12 stg13 stg14 stg15 stg16 stg17"


suggest using extra_image_number to define the additional images, you already support dynamic create images. it is more flexible.

qingwangrh · 2024-12-10T12:11:36Z

qemu/tests/block_hotplug_with_max_map_count.py

+    :param env: Dictionary with test environment
+    """
+
+    process.system("sysctl vm.max_map_count=3072", shell=True)


could you please add reading value and store, then write it back after testing in final block?

qingwangrh · 2024-12-10T12:13:50Z

qemu/tests/block_hotplug_with_max_map_count.py

+    timeout = params.get_numeric("login_timeout", 240)
+    img_list = params.get("images").split()[3:]
+
+    for img in img_list:


for idx in range(extra_image_number):
img="img"+idx
....
you may define the var name you feel good.

mcasquer · 2024-12-11T09:06:57Z

 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: STARTED
 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: PASS (56.46 s)
RESULTS    : PASS 1 | ERROR 0 | FAIL 0 | SKIP 0 | WARN 0 | INTERRUPT 0 | CANCEL 0

mcasquer · 2024-12-11T09:07:12Z

@qingwangrh please could you review again this PR? Thanks !

mcasquer · 2024-12-11T09:08:51Z

@zhencliu @YongxueHong could you also review this PR? Thanks

mcasquer · 2025-01-07T07:02:14Z

Hello @qingwangrh @zhencliu @YongxueHong this is a kindly reminder, could you review this PR? Thanks !

YongxueHong · 2025-01-07T07:41:23Z

qemu/tests/block_hotplug_with_max_map_count.py

+
+def run(test, params, env):
+    """
+    Block hotplug wit vm.max_map_count test


Suggested change

Block hotplug wit vm.max_map_count test

Block hotplug with vm.max_map_count test

good catch, updated !

YongxueHong · 2025-01-07T07:49:44Z

qemu/tests/block_hotplug_with_max_map_count.py

+    test.log.info("Restoring max_map_count original value")
+    process.system(f"sysctl vm.max_map_count={orig_value}", shell=True)


This part should be executed within a finally block, otherwise, the routine can not reach it in any case if it raises an error.

Suggested change

test.log.info("Restoring max_map_count original value")

process.system(f"sysctl vm.max_map_count={orig_value}", shell=True)

finally:

test.log.info("Restoring max_map_count original value")

process.system(f"sysctl vm.max_map_count={orig_value}", shell=True)

mcasquer · 2025-01-08T12:29:21Z

 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: STARTED
 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: PASS (120.18 s)
RESULTS    : PASS 1 | ERROR 0 | FAIL 0 | SKIP 0 | WARN 0 | INTERRUPT 0 | CANCEL 0

mcasquer · 2025-01-08T12:29:54Z

@YongxueHong @qingwangrh @zhencliu could you review again this PR? Thanks !

YongxueHong

LGTM.

qemu/tests/block_hotplug_with_max_map_count.py

qemu/tests/cfg/block_hotplug_with_max_map_count.cfg

mcasquer · 2025-01-10T07:17:04Z

@zhencliu one extra thing, I am not sure if the sysctl command is working since I see invalid argument and the stderr what do you think?

[stdlog] 2025-01-10 02:04:28,831 avocado.utils.process process          L0658 INFO | Running 'sysctl vm.max_map_count=b'3072''
[stdlog] 2025-01-10 02:04:28,836 avocado.utils.process process          L0472 DEBUG| [stderr] sysctl: setting key "vm.max_map_count"
[stdlog] 2025-01-10 02:04:28,836 avocado.utils.process process          L0472 DEBUG| [stderr] : Invalid argument
[stdlog] 2025-01-10 02:04:28,837 avocado.utils.process process          L0715 INFO | Command 'sysctl vm.max_map_count=b'3072'' finished with 0 after 0.001439125s

mcasquer · 2025-01-10T07:17:28Z

 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: STARTED
 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: PASS (57.93 s)
RESULTS    : PASS 1 | ERROR 0 | FAIL 0 | SKIP 0 | WARN 0 | INTERRUPT 0 | CANCEL 0

zhencliu · 2025-01-10T08:49:30Z

@zhencliu one extra thing, I am not sure if the sysctl command is working since I see invalid argument and the stderr what do you think?

[stdlog] 2025-01-10 02:04:28,831 avocado.utils.process process          L0658 INFO | Running 'sysctl vm.max_map_count=b'3072''
[stdlog] 2025-01-10 02:04:28,836 avocado.utils.process process          L0472 DEBUG| [stderr] sysctl: setting key "vm.max_map_count"
[stdlog] 2025-01-10 02:04:28,836 avocado.utils.process process          L0472 DEBUG| [stderr] : Invalid argument
[stdlog] 2025-01-10 02:04:28,837 avocado.utils.process process          L0715 INFO | Command 'sysctl vm.max_map_count=b'3072'' finished with 0 after 0.001439125s

What about trying
count = 3072
system(f"sysctl vm.max_map_count={count}")

mcasquer · 2025-01-10T09:10:48Z

What about trying
count = 3072
system(f"sysctl vm.max_map_count={count}")

Yeah, that works fine :)

Creates new test case that boots up a VM with a small max_map_count, then hotplugs 15 disks and checks the VM status. Signed-off-by: mcasquer <[email protected]>

mcasquer · 2025-01-10T09:19:28Z

@zhencliu code updated ! could you review again the PR? Thanks !

mcasquer · 2025-01-10T09:26:59Z

 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: STARTED
 (1/1) Host_RHEL.m9.u6.ovmf.qcow2.virtio_blk.up.virtio_net.Guest.RHEL.9.6.0.x86_64.io-github-autotest-qemu.block_hotplug_with_max_map_count.q35: PASS (57.26 s)
RESULTS    : PASS 1 | ERROR 0 | FAIL 0 | SKIP 0 | WARN 0 | INTERRUPT 0 | CANCEL 0

mcasquer marked this pull request as ready for review December 2, 2024 14:03

qingwangrh reviewed Dec 10, 2024

View reviewed changes

mcasquer force-pushed the 3130_max_map_count_disk_hotplug branch from b95de9d to 82351a8 Compare December 11, 2024 09:06

YongxueHong reviewed Jan 7, 2025

View reviewed changes

mcasquer force-pushed the 3130_max_map_count_disk_hotplug branch from 82351a8 to 99b965a Compare January 8, 2025 11:43

qingwangrh approved these changes Jan 9, 2025

View reviewed changes

YongxueHong previously approved these changes Jan 9, 2025

View reviewed changes

zhencliu reviewed Jan 9, 2025

View reviewed changes

qemu/tests/block_hotplug_with_max_map_count.py Outdated Show resolved Hide resolved

zhencliu reviewed Jan 9, 2025

View reviewed changes

qemu/tests/cfg/block_hotplug_with_max_map_count.cfg Outdated Show resolved Hide resolved

mcasquer dismissed YongxueHong’s stale review via 88effa6 January 10, 2025 07:14

mcasquer force-pushed the 3130_max_map_count_disk_hotplug branch from 99b965a to 88effa6 Compare January 10, 2025 07:14

block_hotplug_with_max_map_count: creates new test case

cfb8824

Creates new test case that boots up a VM with a small max_map_count, then hotplugs 15 disks and checks the VM status. Signed-off-by: mcasquer <[email protected]>

mcasquer force-pushed the 3130_max_map_count_disk_hotplug branch from 88effa6 to cfb8824 Compare January 10, 2025 09:19

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

block_hotplug_with_max_map_count: creates new test case #4225

block_hotplug_with_max_map_count: creates new test case #4225

mcasquer commented Dec 2, 2024

mcasquer commented Dec 2, 2024

mcasquer commented Dec 2, 2024

mcasquer commented Dec 10, 2024

qingwangrh Dec 10, 2024

mcasquer Dec 11, 2024

qingwangrh Dec 10, 2024

mcasquer Dec 11, 2024

qingwangrh Dec 10, 2024

mcasquer Dec 11, 2024

mcasquer commented Dec 11, 2024

mcasquer commented Dec 11, 2024

mcasquer commented Dec 11, 2024

mcasquer commented Jan 7, 2025

YongxueHong Jan 7, 2025

mcasquer Jan 8, 2025

YongxueHong Jan 7, 2025

mcasquer Jan 8, 2025

mcasquer commented Jan 8, 2025

mcasquer commented Jan 8, 2025

YongxueHong left a comment

mcasquer commented Jan 10, 2025

mcasquer commented Jan 10, 2025

zhencliu commented Jan 10, 2025 •

edited

Loading

mcasquer commented Jan 10, 2025

mcasquer commented Jan 10, 2025

mcasquer commented Jan 10, 2025

	Block hotplug wit vm.max_map_count test
	Block hotplug with vm.max_map_count test

		test.log.info("Restoring max_map_count original value")
		process.system(f"sysctl vm.max_map_count={orig_value}", shell=True)

block_hotplug_with_max_map_count: creates new test case #4225

Are you sure you want to change the base?

block_hotplug_with_max_map_count: creates new test case #4225

Conversation

mcasquer commented Dec 2, 2024

mcasquer commented Dec 2, 2024

mcasquer commented Dec 2, 2024

mcasquer commented Dec 10, 2024

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

mcasquer commented Dec 11, 2024

mcasquer commented Dec 11, 2024

mcasquer commented Dec 11, 2024

mcasquer commented Jan 7, 2025

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

mcasquer commented Jan 8, 2025

mcasquer commented Jan 8, 2025

YongxueHong left a comment

Choose a reason for hiding this comment

mcasquer commented Jan 10, 2025

mcasquer commented Jan 10, 2025

zhencliu commented Jan 10, 2025 • edited Loading

mcasquer commented Jan 10, 2025

mcasquer commented Jan 10, 2025

mcasquer commented Jan 10, 2025

zhencliu commented Jan 10, 2025 •

edited

Loading