Using NVMe SSDs over SATA SSDs in basic servers is an awkward sales pitch
We have traditionally mostlybought basic, inexpensive 1U servers, almost all of which have hadeither two or four drive bays. Our transition of these servers fromusing HDs to using SATA SSDs was relatively straightforward. Itwas driven by dropping SATA SSD prices, the clear improvements inperformance even for casual activities like installing the operatingsystem and upgrading packages (and rebuilding software RAID mirrors),the probable increase in service lifetime, and how relatively easyit was to substitute a SATA SSD for a SATA HD ( except that
My view of Wayland here in 2021
Somewhat recently there was some Wayland related commotionin the Unix tech circles that I vaguely follow, where Wayland peoplewere unhappy about what anti-Wayland people are doing ( for example ,which may have been reacting to this ).Somewhat in reaction to all of this, I had a Twitter reaction :
My Wayland hot take is that I have no idea how well Wayland worksin general, but I do know that for me, switching to it will bevery disruptive because I'll need a new window manager and
A Firefox surprise from disabling dom.event.clipboardevents.enabled
One of the things that browsers allow sites to do with JavaScriptis to intercept and manipulate attempts to copy text out of themor paste text into them. Websites use this to do obnoxious thingslike stop you pasting email addresses and passwords into theirforms or mangling the text you copy out ( via , and this ispotentially a security risk since what you think you're going topaste is not necessarily what you actually get). When I heard aboutthis at some point (I'm not sure when,
Link: "a2d<C-V>3gE: Vim normal mode grammar
Suppose that you're on website B and it wants to send a large chunkof information to some endpoint on website A, through your browser.If this was a small amount of information, website B could use aHTTP redirect with the information stuffed into one or more queryparameters. However, HTTP GET query parameters have a size limitand sometimes you have an especially verbose chunk of informationto transfer. Enter what I've now seen called the "HTTP self-post".In a HTTP self- Sufficiently recent versions of OpenZFS on Linux have not just performance statisticsfor overall pool IO ( also ), but some additional per-datasetIO statistics. Conveniently, these IO statistics are exposed through the Prometheus host agent ,so if you're using Prometheus ( as we are ), so you don't have to writesomething to collect and manipulate them yourself. However, what thesestatistics actually mean is a little bit underexplained. (I believe these first appeared in ZFS 0.8.0, based on In a comment on yesterday's entry on SSD versus NVMe for basic servers , Andrew noted: I object to "SSD vs NVMe" — they're NVMe SSDs. This set me off on a long overdue journey to understand the terminologyhere and get it right. NVMe , also known asNVM Express, is the general standard for accessing non-volatilestorage over PCIe (aka PCI Express ). NVMe doesn't specifyany particular drive form factor or way of physically connectingdrives, but it does mean When I recently wrote about Prometheus's Counting how many times a ((probe_success) time series has startedto fail over a time interval is I was recently reading Russell Coker's Storage Trends 2021 ( via ). As part of the entry, Coker wrote: Last year NVMe prices were very comparable for SSD prices, I washoping that trend would continue and SSDs would go away. [...] Later Coker notes, about the current situation: It seems that NVMe is only really suitable for workstation storage andfor cache etc on a server. So SATA SSDs will be around for a while. Locally, we have an assortment Recently I read A Ceph war story , whichis an alarming story where the day is saved in part by having adetailed hardware inventory that included things like firmwareinformation. This inspired me to think about how we could collect similar informationfor our modest fleet of Ubuntu servers, which led me to "a2d
Learning about the idea of the HTTP self-post
Some things on ZFS (on Linux) per-dataset basic IO statistics
Getting NVMe and related terminology straight (for once)
Counting how many times something started or stopped failing in Prometheus
changes() function and its resets() , I left a larger scale issue not entirelyanswered. Suppose that you have a metric that is either 0 or 1, suchas Blackbox 's probe_success , and you want to know either how many times it'sstarted failing or how many times it's stopped failing over a timeinterval.
SSD versus NVMe for basic servers today (in early 2021)
Vendors put varied and peculiar things in system DMI information
dmidecode , which reports information aboutyour system's hardware as your BIOS describes it according to the SMBIOS / DMI standard(s). When I actuallystarted looking at DMI information for our systems, a number ofinteresting Menu