« first day (1430 days earlier)      last day (3546 days later) » 
00:00 - 19:0019:00 - 00:00

12:41 AM
Speaking of Powershell, when do I get my Parallel.Foreach<IEnumerable> that isn't in a shitty workflow?
 
12:54 AM
Woo our BizSpark application just got approved
 
1:17 AM
@MarkHenderson congratulations. That's a great program.
 
@RyJones Yeah looking forward to utilising it
And getting off of Visual Studio Express
 
heh
 
Might burn through some Azure credits too
Guessing you can't do a huge amount for $160/month but it could be good for some peripheral functionality
 
Bob
> I have backup files that were made in DOS. They contain many hundreds of hours of work, and unfortunately they're locked away in a format that cannot be opened by anything in the format they are now.
...he's asking that in 2014?
 
@Bob We have customer data dating back to 1995 in a production system
 
1:23 AM
I moved some people from DOS to a DOS VM in 2003 or 2004, before they converted to quickbooks in 2005ish
 
Bob
@MarkHenderson Same here, but they're not stored in a totally obsolete backup format.
They're still in the original data files, on a nice shiny NTFS-formatted disk.
 
what format did he use?
 
Bob
2
Q: Backing up and Restoring DOS era backups

user2085722I have backup files that were made in DOS. They contain many hundreds of hours of work, and unfortunately they're locked away in a format that cannot be opened by anything in the format they are now. I have a set of files with extensions .001 through .009. Obviously I can just unzip them, but th...

Some Norton thing, apparently.
 
@Bob o i c
That sucks
 
Bob
Apparently that their only copy of "hundreds of hours of work" is in some ancient backup format :\
I do hope they figure out some way of recovering it, though.
 
1:28 AM
@Bob apparently it wasn't really important, since they didn't need it for a decade+
 
Bob
@RyJones Or their last DOS machine died yesterday? :P
 
@Bob more probable :)
 
1:47 AM
posted on August 11, 2014 by ryan

SSL and TLS have been getting a lot of attention from me lately, and recently I found myself in want of a tool that would tell me precisely which protocol versions and cipher suites a given server supported. Sure, there's SSLScan and SSLScan-Win, but those tools haven't been updated in 5 years, and thus don't support the newer versions of TLS, 1.1 and 1.2.  And of course there are nic

 
All of the employment forms are filled out whew Now off to autozone for paint :(
 
@Jacob Huffing on paint already? And at such a young age. Shame.
 
2:04 AM
@Bob looks like fun. And at least 200 dollars in labour ;p
 
2:31 AM
@mark chip from vehicle shipping
 
@Jacob Hope you can repair it OK. Whenever I've done touchup jobs it looks worse than just leaving it raw
My wife put some nice white scratches in the side of my red car when she scraped a concrete pole a few weeks ago. I buffed out what I could but I know better than to attempt to touch it up myself
 
3:31 AM
@MarkHenderson I do ok, it's mostly to prevent further damage. White hides imperfections well
 
@Jacob white's pretty easy to touch up
 
@RyJones exactly
 
3:51 AM
You know you're in a bad place when your only reasonable choices of database are MySQL and ... Oracle.
 
Bob
@MichaelHampton Wait till you're stuck with Jet (Access) and Excel.
Then you'll be begging for MySQL to save you :P
 
@Bob at one point I think Microsoft was shipping 4 or 5 databases.
Jet, SQL Server, Foxpro, Access (which isn't Jet), and one more I can't remember
 
A relational database management system uses SQL MERGE (also called upsert) statements to INSERT new records or UPDATE existing records depending on whether or not a condition matches. It was officially introduced in the SQL:2003 standard, and expanded in the SQL:2008 standard. == Usage == Right join is employed over the Target (the INTO table) and the Source (the USING table / view / sub-query). That is: If rows present in the Source but missing from the Target do run the action then specifically the NOT MATCHED action If rows missing from the Source and present in Target are ignored then no...
 
 
2 hours later…
5:35 AM
G'day
 
Bob
6:18 AM
o.O
I'm rather surprised to see the ABC here.
 
@Bob ABC advertise on SE quite a bit
Mostly contract positions
I also know a lot of people that have worked on iView
Developer here worked on the iView android port, and another one I used to work with did all the Adobe front-end work for the first iView PC client
 
Bob
@MarkHenderson I never paid attention to those ads.
This one just caught my eye earlier.
Grrrrr imageshack on SE answers
first an adwall
then This image has been deleted
fuck you too
 
6:58 AM
morning
 
7:24 AM
g'day
 
7:39 AM
don't you know it's monday?
 
agreed.. while the fucking moon was that huge tonight there was nearly no sleep. and the heat did not help..
great way to start the week..
 
best monday ever
as I'm still off :)
 
@DennisKaarsemaker you should be shot for that.
 
hi, all. I think this may be the correct place to ask one interview question. please suggest
 
7:59 AM
Just got back from a weeks vacation/wedding in Como, Italy.
That place is magic.. the scenery is breathtaking
 
@pauska as long as you do not need to drive too much in italy i agree.
 
8:15 AM
@Akhil what are you being interviewed for?
Just had some new Dell All-In-Ones arrive...nice. Small screen though, but nice.
 
@tombull89 sorry i am not intervewing anyone. Actually one of my friend went one interview. They ask one question and i also has doubt
Question is Which event is automatically called all time in the gridview?
 
@DennisNolte The driving was OK for me, but I'm from Norway and I'm used to curvy, narrow roads..
 
Not a clue. You'd more likely want a StackOverflow (programmers) room rather than a ServerFault (system administrators) one.
 
@pauska not the roads, the users :)
@tombull89 he did ask at programmers.se and get downvoted because of too broad.
ah its gone already :)
 
I meant ask in a chat room
 
8:22 AM
@tombull89 ah
 
yes. so only checking here. sorry if i am wrong
 
@Akhil why do you think it is server related?
 
@DennisNolte i din't think as server related. I just check and the room is common. so asked here as i didn't found any room related to asp.net or interview
2
 
...
 
gah, this is driving me insane. Still can't figure out why a VM upgraded to from CentOS 5 to 6 has higher I/O load.
 
8:34 AM
@faker did you check per process IO load?
 
8:46 AM
@DennisNolte hm the dirty_ratio changed between 5 and 6. Trying the same ratio now on 6
 
9:01 AM
@faker the usual suspect: increased size of "everything", binaries, libaries. stuff like that.
 
@DennisNolte doubt it, we only run Apache with cronolog (both internally packaged) on it, iotop shows no I/O by anything else
 
morning chaps
 
@faker how much more IO do we talk here? 5/10%?
yo Chooper3
 
@DennisNolte hey, you good?
 
@Chopper3 it is monday..
 
9:11 AM
Today's gem from $other_consultant: "Yes, everything you say is true, but that doesn't mean that you're right and I'm wrong."
7
 
@JennyD impressive.
 
 
@DennisNolte Well, he didn't use those words, but that is the logical conclusion from what he did say
 
ok so factor 10 if i read that correctly.

usual stuff checked i guess (same blocksize, partition mounting, changed default options on certain installed packages.

are you able to use the same versions/kernel for testing?
 
all checked, they are in the same load balancer pool so they get the exact same amount of requests (+-2%).
Can't use the same kernel
 
9:34 AM
did you check the different kernel versions for IO related changes?
does RH have a copy of the kernel config you can diff through?

are you able to testwrite "some " blocks on both devices to actually see if there is a difference in the app ( i am thinking of dd-style writes)
so you can decide if it is underlaying or prog-related
that would be my suggestions on tackling that issue, narrow down the search area :)
i had some IO problems with took me a few weeks to narrow down until i finally figured the issue out, so i am not quite "noob" in that regard, however i am far from an expert though.
@JennyD never conclude anything anyone "says", until you have written proove assume it is spam.
 
@DennisNolte Back-story: He said there was an error in the export of an NFS file system because he couldn't mount it from the server. I replied that this file system is exported to everyone as could be seen by doing showmount -e, but that since it contains home directories, it requires the mounter to have the correct UID for the directory he's trying to mount. And the server he was working on didn't have working LDAP so he didn't have the right UID.
 
@DennisNolte I don't get it really. We only write access logs. Even if scheduling is different the amount of blocks should be roughly the same
 
He replied that this was very true and that we will work on the export permissions tomorrow.
fail
 
@JennyD ah that one, you talked about that the other day , the expert sys admin :)
 
@DennisNolte Yeah, that one... I'm getting more and more certain that I really want a new gig when this year is up.
 
9:48 AM
@faker true that, thats why i trying to say figure out if it is the os or the application(s)
 
10:07 AM
@faker did you build the script to graph that or is it a nagios/icinga script? and if so whats the name, my google fu is bad this morning.
 
10:39 AM
@DennisNolte self written monitor for Xymon
it basically parses iostat output
 
@faker damned :) thats why i could not find it.
 
@DennisNolte it's a really hacky perl script, but it works for us :D
 
@faker we use similiar stuff, though in bash, still "enough" to get an idea if something is wrong or is going wrong.
 
same, since we use Xymon there are pretty much never existing checks. Most of them we implemented in bash too
 
11:31 AM
please kill
-3
Q: Dyman Associates Insurance Group of Companies News: Rika onsdagar - Försäkring tips att komma ihåg

Emil SteimleHar du valt rätt försäkring för dig själv? Sarju Simaria, CFO, Edelweiss Tokio livförsäkring ger dig några punkter att tänka Försäkring är föremålet för värvning. Detta innebär att du som kund göra ett erbjudande som, i stället för en premieinbetalning, livförsäkring företaget skall vid förekoms...

 
@MichaelHampton what the
 
@Nick are you on a mobile connection?
 
@DennisNolte nope, on my desktop
 
@Nick hmm funny, picture is not large (72Kb) so there should not be an issue about that.
 
11:46 AM
refreshing without cache fixed it, weird
 
@Nick i remember things like that on mobile connections, having the fucking providers cache the images, downsize them and then serving them from their server..
some outcry from the ppl who knew what happend, the rest: ignored as usal..
something like this makes me wonder if the programmes and sysadmins of the large companys do not know the law or do not care because they are not responsible?
implementing something like this is cleary illegal (at least in .de) but noone.. cared
 
@DennisNolte its illegal to cache images
?
 
@Nick not the caching, the manipulating.
 
@DennisNolte ah
 
@Nick depending who caches this might be illegal aswell
if the content provider does cache, it is legal (imho) , if it is the provider without agreement it should not be.
 
11:50 AM
Spent 4 hours yesterday trying to install arch linux, only to find that the default DEs suck and need a ton of customization -.-
 
though i am no lawyer, i do see the reasons the ISP do cache, however i did not ask the data from the ISP which might or might not be changed.. but rather from the source i want.
@Nick if you want arch, go for gentoo directly :)
 
one of the reasons HTTPS should be used everywhere. Makes it impossible for that kind of stuff.
 
@DennisNolte why?
@DennisNolte heard some bad stuff about gentoo :p
 
@faker with the kind of CA currently available? and the revoke process? no thx.. fuck ssl..
well, it is at least for me the distribution i learned the most inner setup,dependancies and stuff.
but it is a hell of work.
 
@DennisNolte well what do you recommend, symmetric encryption? :p
 
11:52 AM
@DennisNolte still better then no encryption and allowing anyone to MITM you
 
@Nick encryption? pff, that would mean transfering data over the pipes who actually does matter to someone? no thx. personal messager
@Nick vpn everywhere. get the certs via web of thrust
 
@DennisNolte vpn is not protected outside of the vpn endpoint :p
 
@Nick i did write it so you misunderstood it it seems
example:
example.com is the banking site i want to use
vpn to vpn.example.com, and access "localhost" either directly, or send it via ssh tunnel.
depending on the level you want to use.
performance, throughput, ofc all crap.
 
@DennisNolte why... too much work for nothing, ssl is good if you remove all the CAs you dont like
 
@Nick that would mean all :)
 
11:56 AM
@DennisNolte make your own CA :p
 
then i need to trust the generation process, audit the code etc.
: meh
but i agree, it is better than nothing, but should never be seen as "secure"..
 
@DennisNolte I see you went full paranoid, never go full paranoid
 
@Nick hehe, not quite
@Nick while i do think about stuff like that, i am too lazy for doing it
 
yup same haha
 
though we use a similiar setup for our remote sites.
no external access, ppp connection, vpn from client and then pull the data out.
with the usual stuff like open case warnings etc.
pretty funny stuff to work with, but hard to maintain
 
11:59 AM
it anyway doesn't matter. Google announced that HTTPS sites will get ranked higher starting last week.
Every SEO dude in the world is now telling to you switch to HTTPS ASAP.
 
@faker ye, 1 % impact, fuck that, i could not care less :)
glad i do not need to do SEO :)
 
@DennisNolte 1% now
 
yeye ofc i know
i did read the shit aswell.
in my opinion this means the new NSA datacenter can handle all websites worldwide with HTTPS :)
 
A shop I use advertises that they use 128bit AES HTTPS encryption like its the best thing available -.-
Probably just to impress the people that knows nothing about computers
 
@Nick well, check the creation date, maybe they started advertising this some 10-15 years ago.
and the one responsible got fired, because the page was too slow :)
 
12:02 PM
@DennisNolte the shop opened just like 3 months ago lol
 
@Nick lovely.
great, now i need to learn android fucking library management shit
 
So I read the other day that the NSA was probably behind the DigiNotar hack.
 
possible qoute finfisher support forum:

Very long log in agent
Type: Change Request
Status: Pending
Last update: 2014-06-12 15:32:27
Description: The current log presentation with a long list of loglines dont make any sense to the operator. It would be more useful to trigger the operator on an important error or event, for example by sending an email or SMS text message on fatal errors. With the current logging we missed a disk full warning.
i love it when the operators are too incompetent are actually use surveillance tools
for anyone wanting to read some entries: pastebin.com/RzNz5s9z
 
Came in this morning to a 0x7a BSOD on my work computer ...gonna be one of those types of days apparently
 
12:24 PM
@NathanC ever though of using linux? :p
 
@Nick Unfortunately I can't...I use a lot of Windows-only tools here
 
@NathanC o.O
 
@Nick Virtual Machine 4tw :)
 
True, but my machine doesn't have the RAM to run 2 OS's lol
as it is I bump against the limit with my day-to-day
 
Did anyone else here get asked to participate in a survey for some guys M.Sc. ?
 
12:38 PM
@Iain nope
@NathanC just tell your boss you need new hardware to increase your productivity by X %
 
oh man
We've had cabling contractors in for the past week doing a cable run down to our server run. They've nearly finished and we've just found out they've installed CAT5 instead of CAT6 as specc'd.
 
Its... tooo .... hooot
37 degrees... -.-
 
That's body temperature! What are you complaining about?
 
Thats the room temperature right now...
way too hot for me
let alone be productive with that heat
 
12:54 PM
It's about 19 outside right now. But it's still early in the morning...
 
@tombull89 you try to use 10GBASE-T ?
 
1:32 PM
Sup
 
hey
 
hey @RobM
Can someone go kick one of the guys in our UK office for me?
 
that sounds like a lot of effort
 
as soon as I said that, he IMed me back
>.>
 
maybe they're lurking here O_o
 
1:39 PM
@_@
Honestly, the two guys who take care of most of the server/networking stuff are pretty good
Quiet in here today.
 
well as they're brits they probably stopped answering your IMs because their butler served tea and crumpets. I know that tea and crumpet time is a big drain on productivity here.
3
yes it is actually.
 
I'm ... tuning a database.
 
@cole We're all waiting for Putins next move, in silence
 
EXCUSES!
 
2:29 PM
I'm busy doing nothing at all because Monday
 
2:40 PM
Huh...stupid endpoint deciding a mid-day scan is best apparently...locked me out of everything -_-
 
I love when I actually easily figure something out.
I'll be in Pittsburgh next week.
 
@cole that are the worst things.. at least i allways think i missed some part and it is not that easy, it cannot be, it has never been that easy in the last 1x years..
 
2:57 PM
Horray...crashed with an ntfs.sys error. In safe mode now getting my files backed up before running disk scans :/
 
@NathanC ye, you already said that this day will be bad for you, no need to repeat yourself.
 
hey, Microsoft people... do you see anything wrong with this DNS configuration?
ogc.dom is the internal domain name...
organicgrown.com is the public domain name.
but I'm trying to figure out why an entry like "ppro.organicgrown.com" is its own zone.
 
@ewwhite Is the ppro subdomain internal? They probably did that to "hide" internal IPs
 
no... it's a public domain
Oh, maybe they wanted their public DNS to remain outside.
but still seems off.
 
Yeah...unless they wanted sub-sub domains, it's kinda a waste
 
3:12 PM
SO what I've stumbled upon is a shitfest...
 
Typical for your produce clients, eh?
 
Friday, I went to this client to rack/stack and prep a server
no, this is a place that thinks it's advanced... 4-5 IT people, a manager, director... an outside helpdesk
but everything was fucked up
Pics!
 
Oh, fun.
Apparently I can't run Dell's system detect in safe mode...so probably off to the diag tools for me
 
I went to install my server, and they had no room for it...
"well, we have some space in this rack we zip-tied to the other rack in order to hold batteries. Put your server there"
notices that everything is running off of generic power strips...
despite the rather large APC 8000VA UPS
And I looked at the firewall stack, and it said fuck you!
So I go a little deeper...
Cameras?
No effin' excuse for this.
 
3:38 PM
@ewwhite ahahahaha
How does that even end up happening?
"I'm not sure what's in this rack... but I better label it!"
 
@ewwhite Is that an AP in the rack?
 
@jscott Yes :(
think that pissed me off the most was seeing DNS and vSphere references to the Lefthand SAN.
But it's not a Lefthand SAN. It's an HP P2000 G3 SAN
 
We're having fun today too. :)
Upstairs unit flooded
 
@Tanner plumbing problems?
 
so when I see things like this... mislabeling, bad cabling, weird DNS and domain stuff, total VMware misconfiguration and abysmal uptime, I think it's a good cause to make changes.
@Tanner uh oh
 
3:48 PM
WTF is up with imgur today?
 
@ewwhite Think you need more URLencode in that link there.
 
imgur is fucking up
 
3:59 PM
@jscott Water cooler leaked, or something. I'm not sure.
We had just moved 4 of our IT people to that room last week. Now they have to go somewhere else.
And the construction guys decided to seal up our air conditioner intake for the weekend. Room our servers are in was 98 degrees by Sunday.
 
time for the canonical LACP question
4
Q: FreeBSD link aggregation no faster than single link

Warren YoungWe're trying to use FreeBSD's link aggregation feature in LACP mode to decrease the time it takes to mirror 8 to 16 TiB of data from a master file server to 2-4 clones in parallel. We've got the lagg0 interface configured and working, but it's no faster than the single interface was. That is, a ...

 
4:27 PM
woo I'm two for two good ideas today
 
4:38 PM
@Tanner Mmm...cooking silicon
So, Dell diags say nothing's wrong with the hard drive...so I have no idea.
 
@ewwhite wtf
4 hours ago, by tombull89
We've had cabling contractors in for the past week doing a cable run down to our server run. They've nearly finished and we've just found out they've installed CAT5 instead of CAT6 as specc'd.
 
@NathanC Luckily (unfortunately?) our AS/400 didn't cook.
 
@tombull89 yes?
 
4:53 PM
it's OK I understand what I'm looking at now :P
 
5:17 PM
QQ: Is ServerFault the appropriate place for this question, or should I move it to StackOverflow? serverfault.com/q/619598/237112
 
$ host -t soa yandex.
yandex has SOA record ns1.dns.nic.yandex. hostmaster.neustar.biz. 157 900 900 604800 86400
I wonder how much that cost.
 
@MichaelHampton Wasn't it like half a million?
 
Hm, I don't have that kind of money.
Though, it appears that foo.bar is still available.
 
Plus some largeish annual sum
 
isn't Yandex basically russian Google?
 
5:24 PM
@ArtOfWarfare Seems fine for here - that's more about "how do discover these services" and not "how do I make this code work"
 
@ArtOfWarfare This doesn't look like something that a sysadmin would typically be doing. If you're asking how to code this up or understand the wire protocols you need to use, you're probably better off at Stack Overflow.
 
Sigh.
 
Roar.
 
5:39 PM
Bleugh
 
5:56 PM
Why so quiet here? Are you people all working today?
 
@ewwhite recovering from being blown off by high rollers from Chicago
you show up, say everything compares poorly to Chicago, leave
 
@RyJones Noooo... Seattle > Chicago
But... Portland > Seattle
And I'm typing this while in a mud bath at the Four Seasons hotel spa...
 
ugh, I've spent all day trying to figure out why our upgraded CentOS servers have more I/O. I need a beer now. Or two.
 
@faker I can help that!
 
@ewwhite yes? I saw that on CentOS 6 we have much lower nr_dirty (3200 compared to 34000 on a 5er)
 
6:10 PM
@faker Well, the first thing I tell people is to examine the tuned-adm settings available in EL6.
it's a quick fix to get the type of performance you need, given the use for the server.
3
A: Understanding RedHat's recommended tuned profiles

ewwhiteHere's the schedule of tuned-adm configurations... I think it helps to see them in tabular form. The main thing to note is that the default RHEL6 settings suck!! The other thing is that the enterprise-storage and virtual-guest profiles are identical except for reduced swappiness on the virtual ...

 
@ewwhite and I guess you'd also recommend not using ext3 in CentOS 6?
 
Apply one of those and retest.
I don't recommend ext3... but it happens. I had a few thousand EL6 servers at last job with ext3 because their kickstart build process was never updated to use ext4.
I personally use XFS or ZFS for data partitions.
ext3 is fine... but it all depends on the underlying storage. Are these physical or virtual? RAID controllers? Battery or flash-backed cache?
if any of those, you likely want to remove write barriers.
which again, is kinda handled by the tuned-adm profile
 
I'll try that tomorrow
they are all VMWare VMs on a equallogic SSD SAN
 
@faker tuned-adm profile virtual-guest
Even have a nice snippet of puppet code for it.
# Set appropriate default tuned-adm settings.

class tuned {

  $tuned_pkg = [ "tuned", "tuned-utils", ]
  $tuned_svc = 'tuned'
  $tuned_cmd = 'tuned-adm'

  $tuned_profile = $::virtual ? {
    'vmware'        => 'virtual-guest',
    'physical'      => 'enterprise-storage',
    'vmware_server' => 'default',
    'lxc'           => 'default',
    'default'       => 'default',
  }

  if $::osfamily == 'RedHat' and $::operatingsystemmajrelease > 5 {

    package { $tuned_pkg:
      ensure => "latest",
 
@ewwhite oh nice!
ah, screw it, I try it now. What could possibly go wrong
 
6:21 PM
nothing. you can run this on the fly
 
don't trust him - he's lounging in mud.
 
@RyJones gotta stay pretty :)
 
I wouldn't know, you're too antisocial to mingle.
 
@RyJones Hey... wife + sex.
I was forced to stay in.
 
Since you're a block away, have you tried Japon Dog? yelp.com/biz/gourmet-dog-japon-seattle
Or one of these (widely offered).
 
6:29 PM
@RyJones I walked by it, but was scared away by the people stumbling out of the Canna-bus parked next to the hot dog stand.
Legalized #cannabus #weed #seattle #cannabis http://instagram.com/p/rkPNlMzCjm/
 
I'd try to overcome your provincial attitude toward cannabis consumers to try one, they're great. Also, Uli's in Pike Place has good sausage, although I'm sure it doesn't compare to Chicago's (I'm saying that straight faced, not poking fun)
 
Meh, too much weed around these days. I have friends in the Seattle dispensary business.
but encased meat... I'm all about that!
 
@ewwhite no change :/ it's weird, we get normal amount of writes (100blk/s) for 10 - 20 seconds. Then flush process comes along and writes 20'000blk
 
@ewwhite these things could easily be combined though
 
6:34 PM
@faker How much RAM?
and what type of application is it?
 
@ewwhite 12GB, ~2GB still free. It's Apache+PHP, pretty small. The only writes are access log.
 
@FalconMomot yeah, that is true. I'm trying to see if wife will take a ride with me.
 
@ewwhite I still have one node on CentOS 5 (same load balancer pool, it gets the same # of requests, workload, etc). Comparing graphs:
flat-line vs. spikey-spikey :/
 
what's the impact to the application?
 
6:40 PM
@ewwhite You lined up your arrows. I appreciate your efforts.
 
@ScottPack Puppet Class gave me some vim extensions that do that automatically for you
 
no impact yet, but on vSphere we notice much more I/O usage. We have 50 to 60 of those VMs. Might need to add some storage
 
@ewwhite I saw some but haven't gotten them set up yet. I've mostly relied on puppet-lint
 
@faker hmm... ask a question on main
 
@ewwhite yes, I might. I'll sleep over it.
annoying...
 
6:43 PM
@ScottPack fingers crossed teh peach tree has a dozen or so peaches this year
 
@Iain Awesome!
Any signs of curl?
 
@ScottPack there was bit early on bit I kept it in check
 
Good. I need to have someone come out and see to the magnolia. It's way too much on the house.
 
MY magnolia is small and I can keep ot that way. I need to start managing the oak and the willows some
 
I did just rip out a whole whack of boxwood and replaced them with hosta. Whomever planted them put them on the north side of the house where they got nearly zero sun.
 
6:51 PM
hosta will like that aspect
hosta don't do well here - they become slug food
 
I'm going to keep an eye on the slugs. This summer, and spring for that matter, has been wet and cool.
My cucumber have actually done well this year, that is no slugs, but they're in a much sunnier spot.
 
Our summer has so far been quite wam. I have courgettes (zuccini) that are romping away atm can't eat them fast enough.
I have a Dahlia Imperialis that is over 10' tall it may even flower this year
 
00:00 - 19:0019:00 - 00:00

« first day (1430 days earlier)      last day (3546 days later) »