useless error message

[l8gravely]

When I run: lvcache status data
I get the fairly useless error message of:

need more than 1 value to unpack

So when I do run it with the actual VG/LV I get:

sudo lvcache status data/home
<Command '/sbin/lvs --noheadin...(57 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
<Command '/sbin/lvs --noheadin...(57 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
<Command '/sbin/lvs --noheadin...(57 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
<Command '/sbin/lvs --noheadin...(57 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
<Command '/sbin/lvs --noheadin...(64 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
<Command '/sbin/lvs --noheadin...(68 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
<Command '/sbin/lvs --noheadin...(70 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
<Command '/sbin/lvs --noheadin...(57 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
<Command '/sbin/lvs --noheadin...(54 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
<Command '/sbin/dmsetup status...(22 more)' call_args {'bg': False, 'timeo...(476 more)>: starting process
+-----------------------+-------------------+
| Field | Value |
+-----------------------+-------------------+
| cached | True |
| size | 590558003200 |
| cache_lv | homecacheLV |
| cache_lv_size | 53687091200 |
| metadata_lv | homecacheLV_cmeta |
| metadata_lv_size | 134217728 |
| cache_block_size | 128 |
| cache_utilization | 819200/819200 |
| cache_utilization_pct | 100.0 |
| demotions | 4024 |
| dirty | 0 |
| end | 1153433600 |
| features | 1 |
| md_block_size | 8 |
| md_utilization | 2443/32768 |
| md_utilization_pct | 7.45544433594 |
| promotions | 4024 |
| read_hits | 6214751 |
| read_misses | 2514189 |
| segment_type | cache |
| start | 0 |
| write_hits | 22841584 |
| write_misses | 3841539 |
+-----------------------+-------------------+

Which is fine, but the extra output is annoying when I haven't asked for verbose. I'd also like it to just find ALL LVs in my VG and display them, if it can.

[romanrm]

I get the same issues, except I don't even get the actual output table. The last lines of output are:

<Command '/sbin/dmsetup status /dev/mapper/test-nbd7'>: starting process
<Command '/sbin/dmsetup status /dev/mapper/test-nbd7', pid 21925>: process started
<Command '/sbin/dmsetup status /dev/mapper/test-nbd7', pid 21925>: process completed
<Command '/sbin/dmsetup status /dev/mapper/test-nbd7', pid 21925>: process completed
<Command '/sbin/dmsetup status /dev/mapper/test-nbd7', pid 21925>: process completed
dictionary changed size during iteration

[ss23]

@romanrm Fixed with PR #14

[l8gravely]

Well, I had some time and I chased down this issue a little bit. First off the 'sh' module in python is what generates the '<Command ....' lines, and it defaults to clamping the output to 20 characters for the command. Which is useless for debugging purposes. I manually hacked the version I have to up the limit, but that's just a hack.

Second, the code assumes you give it / for alot of the commands, but doesn't really tell you that nicely. It just needs to be updated to have better error messages.