The Chubby Lock Service for Loosely-Coupled Distributed Systems

• p. 8 §2.9, Figure 2 walk-through

  wrong mas-ter epoch number (described in detail below)

• p. 9 §2.9, step 1

  It first picks a new client epoch number

Judge's note: Text checked on p.8 line 884 and p.9 lines 908–938, against the full page images. The forward reference points to step 1, which calls the number a 'client epoch number'. It is the same number, chosen by the master and presented by clients, so either name can be defended. Still, the change of name could lead a reader to think there are two counters.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 10 §3, paragraph before §3.1

  We do not yet use them in production

• p. 11 §4.1 table

  additional proxied clients

Only the authors can say what is right here.

Judge's note: Checked §3 on p.10 and the §4.1 table on p.11. The paper does not define 'proxied clients' in the table as clients served through protocol-conversion servers rather than Chubby proxies. §4.3 ("might have chosen to implement full proxies sooner than we did") and §4.4 (redesign of fail-over for proxies) also suggest proxies exist. As published, the statements are in tension, and only the authors can say what the 32k counts.

Second check: another model agreed

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 2 §2.1, third argument, right column

  think they know to use

Judge's note: Page 2 crop, right column, shows "think they know to use / them". The word 'how' is missing.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 4 §2.3, left column

  The design differs from UNIX in a ways that ease dis-

Judge's note: The page 4 crop shows "in a ways that ease dis-tribution". The singular article 'a' cannot go with plural 'ways'.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 6 §2.6 API, second paragraph

  the library provides a handle on ”/” that is always

Judge's note: The page 6 crop shows both quote marks around / in closing (right-hand) form. Elsewhere the paper uses proper opening and closing quotes, e.g. “up” on p.11.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 8 §2.9, right column

  both as viewed by both the old and new masters

Judge's note: The page 8 crop, right column, shows 'both as viewed' / 'by both the old and new masters'.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 9 §2.10, right column

  this allows nodes to by keyed by their path name

Judge's note: The page 9 crop, right column, confirms 'to by keyed'.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 9 §2.10, second paragraph

  Berkeley DB’s uses a distributed consensus protocol

Judge's note: The page 9 crop, right column, confirms 'Berkeley DB’s uses'.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 10 Sec 3

  (see 4.1)

Judge's note: The page 10 crop confirms '(see 4.1)' with no § symbol, so this is not an extraction loss. The same page uses '(see §4.1)' in §3.1.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 11 §4.1, first sentence

  the RPC rate was a seen over a ten-minute

Judge's note: The page 11 crop confirms 'was a seen'.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 12 §4.3, right column, top

  several jobs many be running at once

Judge's note: The page 12 crop, right column, confirms 'many be running'.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 14 §4.5 Publish/subscribe, top left

  make it a slow and inef-

• p. 14

  make it a slow and inef-ficient for all but the most trivial

Judge's note: The page 14 crop, left column, confirms 'make it a slow and inef-ficient for all but'.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 14 §4.6 Developers rarely consider availability

  employing hundred of machines

Judge's note: The page 14 crop, left column, confirms 'employing hundred of machines'.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 14 §4.6, left column

  has little or no affect on their

• p. 14 Sec 4.6 heading

  Poor API choices have unexpected affects

Judge's note: The page 14 crops confirm both uses of 'affect' as a noun: one in the left column body text, one in the right column bold run-in heading.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 14 §4.6 bold paragraph heading

  Poor API choices have unexpected affects

Judge's note: I checked the crop of the page-14 right column. The bold heading reads 'Poor API choices have unexpected affects'. A noun is needed here, so 'effects' is the correct word. The same slip appears in the left column ('little or no affect on their applications'), which the authors may want to fix at the same time.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 14 §4.6 RPC use affects transport protocols

  used in the same way KeepAlives.

Judge's note: The crop of the page-14 right column shows '...in the normal case, used in the same way KeepAlives. The KeepAlive reply...'. The phrase falls on a single line, so a line break does not explain it. 'as' is missing.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 15 §5, left column, bottom

  the lock and file names spaces

Judge's note: Page 15, bottom of the left column, reads 'Chubby combines the lock and file names spaces'. Elsewhere the paper writes 'name space' (for example, on the same page: 'a file-system-like name space'). This is a typo.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 15 §5, last paragraph

  need sequencers to allow externals resources

• p. 15 Section 5, final paragraph

  need sequencers to allow externals resources to be pro-

• p. 15 Sec 5

  allow externals resources to be pro-

Judge's note: Page 15, right column, last paragraph of §5, reads 'allow externals resources to be pro-tected safely'. The word is used as an adjective, so it should be 'external'.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 15 §6 Summary

  simple load adaptation to allow it scale to tens of thou-

• p. 15 Sec 6

  to allow it scale to tens of thou-

Judge's note: Page 15, §6, second paragraph, reads 'simple load adaptation to allow it scale to tens of thou-sands of client processes'. The word 'to' is missing.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send

• p. 16 Reference [11]

  ACM TOPLAS, 3 (1985),

Only the authors can say what is right here.

Judge's note: Rests on outside knowledge: the correct volume number for this TOPLAS paper is not in the files. The page-16 reference crop shows '[11] JEFFERSON, D. Virtual time. ACM TOPLAS, 3 (1985), 404–425.' The other journal entries give the volume before the issue: [10] 'ACM TOCS 6, 1', [12] 'ACM TOCS 16, 2', [13] 'ACM SIGACT News 32, 4'. In [11] the volume is missing. The fact that it is missing can be seen on the page, but the correct number has to come from the authors or the source, so this is an author query.

Second check: none (judge only)

Report a problem with this finding

This finding is

wrong right unclear or overstated already corrected (erratum, later version)

I am an author of this paper Send