Excellent...thanks.

On Thu, 21 Dec 2000 18:03:29 -0500 (EST), rpjday wrote:

>On Thu, 21 Dec 2000, rpjday wrote:
>
>> On Thu, 21 Dec 2000, Fred Edmister wrote:
>>
>> >          From my experience, the +'s mean that when the system was
>> > configured, those partitions were set to expand to fill the disk should
>> > there be any room left...  I have one partition set on my servers to
>> > expand, I'm not sure what the details are on the expansion... (wether it
>> > only expands as needed or just when there is extra space, it automatically
>> > claims it.. )  Hope this helps.
>> >
>> nope, that ain't it.  first, why should there be *any* connection
>> between making a partition growable during a linux install and what
>> kind of attributes that partition gets on the disk afterwards?
>>
>> second, i just checked, and an explicitly-sized /boot partition
>> on one of my hosts has the "+".
>>
>> i still think it has to do with not being on an even cylinder
>> boundary.
>
>ok.  when in doubt, read the source.  just grabbed the util-linux
>RPM and installed it, zipping thru fdisk.c (if you'd like to follow
>along, the salient code starts at line 1388, and contains the
>following excerpt:
>
>
>
>       printf(_("%*s Boot    Start       End    Blocks   Id  System\n"),
>              w+1, _("Device"));
>
>       for (i = 0 ; i < partitions; i++) {
>               struct pte *pe = &ptes[i];
>
>               p = pe->part_table;
>               if (p->sys_ind) {
>                       unsigned int psects = get_nr_sects(p);
>                       unsigned int pblocks = psects;
>                       unsigned int podd = 0;
>
>                       if (sector_size < 1024) {
>                               pblocks /= (1024 / sector_size);
>                               podd = psects % (1024 / sector_size);
>                       }
>
>... some stuff deleted ...
>
>/* odd flag on end */  (long) pblocks, podd ? '+' : ' ',   <== AHA!!!!!
>/* type id */          p->sys_ind,
>/* type name */                (type = partition_type(p->sys_ind)) ?
>                       type : _("Unknown"));
>                       check_consistency(p, i);
>               }
>       }
>
>  so a '+' is printed when the partition is not an even number of
>*logical* blocks.
>
>rday
>
>
>
>_______________________________________________
>Redhat-list mailing list
>[EMAIL PROTECTED]
>https://listman.redhat.com/mailman/listinfo/redhat-list
>





_______________________________________________
Redhat-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/redhat-list

Reply via email to