Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed? • Hacker News Board

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#131

Post by tcpekin » Fri, Sep 30, 2022, 4:23 PM UTC

Mac M1 Air | 16 GB RAM | 1TB SSD | Purchased Dec 2021, used daily since

    === START OF SMART DATA SECTION ===
    SMART overall-health self-assessment test result: PASSED

    SMART/Health Information (NVMe Log 0x02)
    Critical Warning:                   0x00
    Temperature:                        35 Celsius
    Available Spare:                    100%
    Available Spare Threshold:          99%
    Percentage Used:                    4%
    Data Units Read:                    268,644,677 [137 TB]
    Data Units Written:                 259,969,878 [133 TB] 
    Host Read Commands:                 1,213,441,896
    Host Write Commands:                981,264,397
    Controller Busy Time:               0
    Power Cycles:                       125
    Power On Hours:                     793
    Unsafe Shutdowns:                   7
    Media and Data Integrity Errors:    0
    Error Information Log Entries:      0

    Read 1 entries from Error Information Log failed: GetLogPage failed: system=0x38, sub=0x0, code=745

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#132

Post by neurostimulant » Fri, Sep 30, 2022, 4:37 PM UTC

jackjeff wrote: post #76

Mac Studio M1 Max / 32 GB RAM / 2TB SSD / April 2022 Used daily for moderate "office" usage (Slack/Teams/Firefox/RDP) during office hours. Heavier usage (XCode/AndroidStudio/Lightroom/FCP) on weekends/evenings. === START OF SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED SMART/Health Information (NVMe Log 0x02) Critical Warning: 0x00 Temperature: 27 Celsius Available Spare: 100% Availa…

Wow. My primary SSD on linux (with 32gb ram) has 2598 hours uptime logged and only 2.4 TB written so far.

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#133

Post by neurostimulant » Fri, Sep 30, 2022, 4:52 PM UTC

unfocused wrote: post #33

M1 Air 16G 1TB (Bought on Nov 2020 - 1y 10 months old) I think I'm second on this page with "unsafe shutdowns" with 62. It's used by 4 people, 2 adults, 2 kids and heavily used pretty much all day and night. Lots of web, and video watching. Some very light development and a ton of email :) I think the unsafe shutdowns are due to my kids trying to run Steam and other games that may or may not work properly, e.g. Farmi…

This is a lot lighter usage than what other people posted here. Are you rarely opening many apps at once until your ram almost full?

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#134

Post by cool_hw » Fri, Sep 30, 2022, 5:19 PM UTC

tcpekin wrote: post #131

Mac M1 Air | 16 GB RAM | 1TB SSD | Purchased Dec 2021, used daily since === START OF SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED SMART/Health Information (NVMe Log 0x02) Critical Warning: 0x00 Temperature: 35 Celsius Available Spare: 100% Available Spare Threshold: 99% Percentage Used: 4% Data Units Read: 268,644,677 [137 TB] Data Units Written: 259,969,878 [133 TB] Host Read Comma…

How does your workload look?

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#135

Post by cool_hw » Fri, Sep 30, 2022, 5:21 PM UTC

neurostimulant wrote: post #132

jackjeff wrote: post #76

Mac Studio M1 Max / 32 GB RAM / 2TB SSD / April 2022 Used daily for moderate "office" usage (Slack/Teams/Firefox/RDP) during office hours. Heavier usage (XCode/AndroidStudio/Lightroom/FCP) on weekends/evenings. === START OF SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED SMART/Health Information (NVMe Log 0x02) Critical Warning: 0x00 Temperature: 27 Celsius Available Spare: 100% Availa…

Wow. My primary SSD on linux (with 32gb ram) has 2598 hours uptime logged and only 2.4 TB written so far.

This is exactly the kind of comparison that prompted me to post this topic. I suppose macOS just uses the SSD far more.

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#136

Post by doerig » Fri, Sep 30, 2022, 5:50 PM UTC

2016 MBP, reaching its end of life. Power cycles and unsafe shutdowns are worrying, not to mention the failed S.M.A.R.T. status. It has become almost unusable the last few months. I will purchase the new M2 MBP as soon as they release it.

    === START OF SMART DATA SECTION ===
    SMART overall-health self-assessment test result: FAILED!
    - NVM subsystem reliability has been degraded

    SMART/Health Information (NVMe Log 0x02)
    Critical Warning:                   0x04
    Temperature:                        29 Celsius
    Available Spare:                    100%
    Available Spare Threshold:          10%
    Percentage Used:                    134%
    Data Units Read:                    646,194,436 [330 TB]
    Data Units Written:                 642,615,704 [329 TB]
    Host Read Commands:                 6,646,415,427
    Host Write Commands:                5,882,338,165
    Controller Busy Time:               34,784
    Power Cycles:                       86,652
    Power On Hours:                     4,505
    Unsafe Shutdowns:                   1,060
    Media and Data Integrity Errors:    0
    Error Information Log Entries:      932

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#137

Post by simfree » Sat, Oct 01, 2022, 7:28 AM UTC

doerig wrote: post #136

2016 MBP, reaching its end of life. Power cycles and unsafe shutdowns are worrying, not to mention the failed S.M.A.R.T. status. It has become almost unusable the last few months. I will purchase the new M2 MBP as soon as they release it. === START OF SMART DATA SECTION === SMART overall-health self-assessment test result: FAILED! - NVM subsystem reliability has been degraded SMART/Health Information (NVMe Log 0x02)…

Replace the SSD asap. It's not worth saving $80 to deal with the random IO lag well worn SSDs have.

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#138

Post by doerig » Sat, Oct 01, 2022, 7:54 AM UTC

simfree wrote: post #137

doerig wrote: post #136

2016 MBP, reaching its end of life. Power cycles and unsafe shutdowns are worrying, not to mention the failed S.M.A.R.T. status. It has become almost unusable the last few months. I will purchase the new M2 MBP as soon as they release it. === START OF SMART DATA SECTION === SMART overall-health self-assessment test result: FAILED! - NVM subsystem reliability has been degraded SMART/Health Information (NVMe Log 0x02)…

Replace the SSD asap. It's not worth saving $80 to deal with the random IO lag well worn SSDs have.

Forgot to specify that I have the 15" model. So I would have to replace the entire main board since the SSD is soldered onto it...

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#139

Post by frankjr » Sat, Oct 01, 2022, 9:12 AM UTC

cool_hw wrote: post #115

frankjr wrote: post #108

Earlier quoted context omitted.

I'm surprised by the numbers here as well. I have a late 2016 MBP 512GB (used daily) and the numbers are nowhere similar. It's a bit worrying though that I'm the only one in this thread with error log entries. SMART/Health Information (NVMe Log 0x02) Critical Warning: 0x00 Temperature: 30 Celsius Available Spare: 100% Available Spare Threshold: 10% Percentage Used: 3% Data Units Read: 59,731,850 [30.5 TB] Data Units…

Is there a way to look up that error log?

  Error Information (NVMe Log 0x01, 16 of 64 entries)
  Num   ErrCount  SQId   CmdId  Status  PELoc          LBA  NSID    VS
    0         16     0  0x0000  0x4016  0x004            0     1     -
    1         15     0  0x0000  0x4016  0x004            0     1     -
    2         14     0  0x0000  0x4016  0x004            0     1     -
    3         13     0  0x0000  0x4016  0x004            0     1     -
    4         12     0  0x0000  0x4016  0x004            0     1     -
    5         11     0  0x0000  0x4016  0x004            0     1     -
    6         10     0  0x0000  0x4016  0x004            0     1     -
    7          9     0  0x0000  0x4016  0x004            0     1     -
    8          8     0  0x0000  0x4016  0x004            0     1     -
    9          7     0  0x0000  0x4016  0x004            0     1     -
   10          6     0  0x0000  0x4016  0x004            0     1     -
   11          5     0  0x0000  0x4016  0x004            0     1     -
   12          4     0  0x0000  0x4016  0x004            0     1     -
   13          3     0  0x0000  0x4016  0x004            0     1     -
   14          2     0  0x0000  0x4016  0x004            0     1     -
   15          1     0  0x0000  0x4016  0x004            0     1     -
  ... (48 entries not read)

To be honest I'm not sure how to interpret it. There are some tools on Linux to read this information but the output doesn't really seem more useful.

Re: Ask HN: Has the Apple Silicon excessive disk read/write issue been fixed?

#140

Post by cool_hw » Sat, Oct 01, 2022, 11:14 AM UTC

frankjr wrote: post #139

cool_hw wrote: post #115

Earlier quoted context omitted.

Is there a way to look up that error log?

Error Information (NVMe Log 0x01, 16 of 64 entries) Num ErrCount SQId CmdId Status PELoc LBA NSID VS 0 16 0 0x0000 0x4016 0x004 0 1 - 1 15 0 0x0000 0x4016 0x004 0 1 - 2 14 0 0x0000 0x4016 0x004 0 1 - 3 13 0 0x0000 0x4016 0x004 0 1 - 4 12 0 0x0000 0x4016 0x004 0 1 - 5 11 0 0x0000 0x4016 0x004 0 1 - 6 10 0 0x0000 0x4016 0x004 0 1 - 7 9 0 0x0000 0x4016 0x004 0 1 - 8 8 0 0x0000 0x4016 0x004 0 1 - 9 7 0 0x0000 0x4016 0x00…

I did a quick search earlier and it seems these are informational only and not necessarily persistent warnings. Some discussion suggested that some unsupported requests to the disk can result in these, but I don't know for sure.