Attribute width not allowed on element input at this point.
<input type=number id=gotoline width=6>
Width attribute is only valid on input element for type image.
Changed to use size attribute.
Stray end tag input.
Bad value for attribute cols on element textarea: The empty string is not a valid positive integer.
<textarea id=fileContent name=fileContent class=form-control rows=30 cols=">
Syntax of positive integer: One or more digits (0-9), with at least one which is non-zero.
For example: 42 is valid, but 00 is not.
This is being handled by css so setting to default value.
Stray start tag script.
</html>↩↩<script type=text/javascript>↩$(doc
Scripts being placed after the page closing tags is bad form.
The scripts in foot.inc that are required by filebrowser/browse.js should be provide by other means such as a scripts include file so that browser.js, and others, can be placed before the page close tags of foot.inc. Foot.inc should be just that. Not the start of something more. This is a design issue. Leaving for another time.
1) Add code from RELENG_2_2 cf3aff59ed that checks for leases that are in pools other than the ordinary first pool. This code has been missed from the bootstrap conversion.
2) Make the leases per interface/pool work properly. The counter $dlsc did not get incremented much anyway, because of break statements in the previous code. I have made it key $dhcp_leases_subnet_counter by the interface and starting IP of the pool - this is a unique identifier of each pool, so the counter for it can be reliably incremented as leases are considered in any order.
3) Sort $dhcp_leases_subnet_counter before displaying the counter data - that way the displayed data is always nicely ordered by interface name and then increasing pool start IP, regardless of the order that the leases happened to be found.
Remove "$entry['iaid'] . " from $mappings[] since the IAID of the IA_NA != IAID of the IA_PD, please see note in https://redmine.pfsense.org/issues/4206#note-9
Add table header for the icon in the Delegated Prefixes section to fix alignment of the column headers.
The last test here looks not good having just a single equals sign - if the interpreter evaluates this then $class will always get set to 'info'.
In practice it should work OK, because the evaluation of boolean conditions left to right should stop when any of the other tests to the left evaluates false. So $class will only be set to 'info' if it is not already 'default' 'danger' or 'warning'.
Actually the code could be:
a) Obfuscated version with empty if block content:
if ($class != 'default' && $class != 'danger' && $class != 'warning' && $class = 'info') {
// Noop - the if test above has already set $class to 'info' as required.
}
or
b) Clearer version without bothering with any test for $class != 'info' :
if ($class != 'default' && $class != 'danger' && $class != 'warning') {
$class = 'info';
}
Forum: https://forum.pfsense.org/index.php?topic=106547.0
The underlying FreeBSD command seems to allocate space in MiB - e.g. https://www.freebsd.org/doc/handbook/disks-virtual.html examples show a 5m disk getting 10240 sectors (of 512 bytes) = 5120 KiB = 5 MiB. Even though the formatted space shown by df ends up being less, the underlying total allocated space is the specified size in MiB.
/etc/rc.embedded does commands like:
mdmfs -S -M -s 123m md /tmp
Which will make a 123MiB memory disk.
