pt-summary fusionIO Use of uninitialized value $tmp_adapter in concatenation (.) or string

Bug #1240534 reported by Kenny Gryp
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Percona Toolkit moved to https://jira.percona.com/projects/PT
Confirmed
Undecided
Unassigned

Bug Description

on a fusionIO device, pt-summary is stuck, waiting for nothing.

However, the collect.err shows an error.

Here's how I reproduce it:

root@mysql01:~/percona/bin# perl -wln /tmp/pt-summary.13297.zOBcze/fio_status_format.pl /tmp/pt-summary.13297.zOBcze/data/fusion-io_card_original_output
driver_version 2.3.11 build 183
Use of uninitialized value $tmp_adapter in concatenation (.) or string at /tmp/pt-summary.13297.zOBcze/fio_status_format.pl line 12, <> chunk 31.

contents of fusion-io_card_original_output:

Found 1 ioDrive in this system
Fusion-io driver version: 2.3.11 build 183

fct0 Attached as 'fioa' (block device)
 HP StorageWorks 320GB IO Accelerator, Product Number:AJ878B SN:603047
 HP Mezz3 320GB, PN:00242123456, Mfr:003, Date:20120522
 Alt PN:123456-001
 Powerloss protection: protected
 PCI:06:00.0
 Vendor:1aed, Device:1003, Sub vendor:103c, Sub device:324d
 Firmware v5.0.7, rev 107053
 320.00 GBytes block device size, 406 GBytes physical device size
 Format: block, v300, 625,001,920 sectors, 512 bytes per sector
 Error correction: 39 bits per 960 bytes
 FPGA ID:0 Format UID:0000000933a701331234567890123456
 PCIE slot available power: unavailable
 Sufficient power available: Unknown
 PCIE negotiated link: 4 lanes at 2.50 Gbits/sec each, 1000 MBytes/sec total
 Internal temperature: 53.6 degC, max 58.1 degC
 Board temperature: 47 degC
 Internal voltage: avg 1.022V, max 1.025V
 Aux voltage: avg 2.464V, max 2.467V
 Media status: Healthy; Reserves: 100.00%, warn at 10.00%
 Lifetime data volumes:
    Physical bytes written: 321,473,910,107,848
    Physical bytes read : 237,597,764,591,072
 RAM usage:
    Current: 131,184,640 bytes
    Peak : 165,382,144 bytes

Tags: pt-summary
Revision history for this message
Kenny Gryp (gryp) wrote :

./pt-summary --version
local: 1: 2.2.4: bad variable name
pt-summary

tags: added: pt-summary
Changed in percona-toolkit:
status: New → Confirmed
Revision history for this message
IT Operations (itops-box) wrote :

same problem for 2.2.6, different type of fusion-io card. This is with the following fio driver/tools versions:
fio-util-2.3.11.183-1.0.el6.x86_64
fio-firmware-107053-1.0.noarch
fio-common-2.3.11.183-1.0.el6.x86_64
libfio-2.3.11.183-1.0.el6.x86_64
fio-sysvinit-2.3.11.183-1.0.el6.x86_64

Would those be newer or older than what pt-summary expects?

I also noticed that I could not work around the problem by doing
pt-summary --summarize-network
it would still do the fio work and hang on it.

My fio card output that pt-summary is trying to parse:
Found 2 ioDrives in this system
Fusion-io driver version: 2.3.1 build 123

        Attached as 'fioa' (block device)
        Fusion-io ioDrive Duo 1.28TB, Product Number:24X4P SN:460424
        ioDIMM3 640GB MLC, PN:00278400602, Mfr:003, Date:20111121
        PCI:07:00.0
        Vendor:1aed, Device:1005, Sub vendor:1028, Sub device:1f66
        Firmware v5.0.7, rev 107053
        640.00 GBytes block device size, 812 GBytes physical device size
        Format: block, v300, 1,250,001,920 sectors, 512 bytes per sector
        Error correction: 39 bits per 960 bytes
        FPGA ID:0 Format UID:0000000706880132df11004260396800
        PCIE slot available power: 25.00W
        PCIE negotiated link: 4 lanes at 2.50 Gbits/sec each, 1000 MBytes/sec total
        Internal temperature: 44.8 degC, max 49.2 degC
        Board temperature: 38 degC
        Internal voltage: avg 1.028V, max 1.034V
        Aux voltage: avg 2.473V, max 2.473V
        Media status: Healthy; Reserves: 100.00%, warn at 10.00%
        Lifetime data volumes:
           Physical bytes written: 1,161,834,893,547,768
           Physical bytes read : 1,229,596,674,965,056
        RAM usage:
           Current: 343,046,144 bytes
           Peak : 358,840,320 bytes

        Attached as 'fiob' (block device)
        Fusion-io ioDrive Duo 1.28TB, Product Number:24X4P SN:460422
        ioDIMM3 640GB MLC, PN:00278400602, Mfr:003, Date:20111121
        PCI:08:00.0
        Vendor:1aed, Device:1005, Sub vendor:1028, Sub device:1f66
        Firmware v5.0.7, rev 107053
        640.00 GBytes block device size, 812 GBytes physical device size
        Format: block, v300, 1,250,001,920 sectors, 512 bytes per sector
        Error correction: 39 bits per 960 bytes
        FPGA ID:0 Format UID:0000000706860132df11004260396800
        PCIE slot available power: 25.00W
        PCIE negotiated link: 4 lanes at 2.50 Gbits/sec each, 1000 MBytes/sec total
        Internal temperature: 41.8 degC, max 43.8 degC
        Board temperature: 32 degC
        Internal voltage: avg 1.011V, max 1.020V
        Aux voltage: avg 2.467V, max 2.467V
        Media status: Healthy; Reserves: 100.00%, warn at 10.00%
        Lifetime data volumes:
           Physical bytes written: 909,374,648,029,536
           Physical bytes read : 914,990,620,713,232
        RAM usage:
           Current: 283,437,056 bytes
           Peak : 294,049,792 bytes

Revision history for this message
Shahriyar Rzayev (rzayev-sehriyar) wrote :

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PT-1161

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.