Bug 1622616
Summary: | Inconsistency over fio utility benchmarking | ||
---|---|---|---|
Product: | Red Hat Satellite | Reporter: | anerurka |
Component: | Satellite Maintain | Assignee: | Kavita <kgaikwad> |
Status: | CLOSED WONTFIX | QA Contact: | Nikhil Kathole <nkathole> |
Severity: | medium | Docs Contact: | |
Priority: | medium | ||
Version: | 6.4 | CC: | anerurka, apatel, aupadhye, dgupte, francesco.trentini, gerben, inecas, johannes.grumboeck, kgaikwad, ktordeur, kupadhya, mark.jackson2, mbacovsk, peter.vreman, rvdwees, sitsofe, sshtein |
Target Milestone: | Unspecified | Keywords: | Performance, Triaged |
Target Release: | Unused | ||
Hardware: | x86_64 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | If docs needed, set a value | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2020-04-02 12:32:34 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | |||
Bug Blocks: | 1122832, 1619394 |
Description
anerurka
2018-08-27 15:33:15 UTC
Created redmine issue http://projects.theforeman.org/issues/24959 from this bug Upstream bug assigned to kgaikwad (Random passerby comment) At least the first fio job listed here (fio --name=job1 --rw=read --size=1g --directory=/var/lib/pulp --direct=1) has the same issue as described over in bug 1641784 comment 12 - you're reading synchronous I/O where you're ONLY sending a 4KByte block down as direct I/O and waiting for it to come back before sending any more. You're bypassing the Linux page cache because you asked for direct I/O (so no coalescing of those tiny I/Os, you won't get/benefit from readahead). You aren't sending I/O down in parallel. Are you sure something like a spinning disk can really do more than 24 MBytes a second with such tight restraints? > Based on the above results it looks like, fio test can not be trusted for real workload simulation. That's an extreme statement which might need some qualification words around it :-) Perhaps the fio jobs being requested haven't been fully understood? Maybe someone could sit down and take in the huge range of options and caveats mentioned over in the fio help - https://fio.readthedocs.io/en/latest/fio_doc.html and chat with your Linux (disk I/O) folks about these affect kernel submission... The tool provided for pre upgrade checks also reported this for myself. I was able to get results that were in line with what I would expect by using a bigger block size and/or more threads. The problem however, is that the upgrade tool and documentation [https://access.redhat.com/solutions/3397771] are misleading. Looking at the documentation you are told to run 'fio --name=job1 --rw=read --size=1g --directory=/var --direct=1'. If something else should be run, the documentation, and the upgrade check tool (assuming it is just running fio), should also be updated. The Satellite Team is attempting to provide an accurate backlog of bugzilla requests which we feel will be resolved in the next few releases. We do not believe this bugzilla will meet that criteria, and have plans to close it out in 1 month. This is not a reflection on the validity of the request, but a reflection of the many priorities for the product. If you have any concerns about this, feel free to contact Red Hat Technical Support or your account team. If we do not hear from you, we will close this bug out. Thank you. Thank you for your interest in Satellite 6. We have evaluated this request, and while we recognize that it is a valid request, we do not expect this to be implemented in the product in the foreseeable future. This is due to other priorities for the product, and not a reflection on the request itself. We are therefore closing this out as WONTFIX. If you have any concerns about this, please do not reopen. Instead, feel free to contact Red Hat Technical Support. Thank you. |